view additional pages

Video

COMPUTERS AND VIDEO appear to offer an endless variety of combinations, as this month’s cover by Robert Tinney depicts. With an increase in higher-power communication satellites that require smaller, less-expensive user antennas and electronics, and with the melding of television receivers and microprocessors, we might look ahead to the day when worldwide person-to-person visual as well as aural communication is based on personal computers and not on a direct descendant of Mr. Bell’s original invention. “Bulletin Boards in Space.” described by John Markoff in the May BYTE West Coast column (page 88), may give way to worldwide electronic conferencing and conventioning by adding television cameras to the growing list of common personal computer user options.

Many videotex, work-alike, and other interactive cable-television-based systems already exist. With the proliferation of coaxial-cable interconnection and its high-speed data-transmission capabilities, more and more homes and businesses could be linked via computer-controlled video instead of the restricted-bandwidth, audio-frequency-based systems now in use.

Shopping via computer is already a reality in some areas. With television- or presentation-level graphics, perhaps complicated and expensive encoding schemes could become unnecessary because you might be able to view the person to whom your message is directed. And as Rich Malloy, our product-review editor, stated in the July 1983 BYTE in his introduction to the Videotex theme, the printing presses might stop and BYTE could be delivered to you electronically.

Though with us for over a generation, until recently hardware costs made computer and video interactivity limited and expensive. In recent years, dramatic advances in digital electronics and large-scale integration (LSI) have made personal computers, videocassette recorders (VCRs), and videodiscs available to anyone with a down-to-earth application or interest in learning. In this issue, we present articles on a wide range of topics related to computers and video.

Peter R. Cook’s article, “Electronic Encyclopedias.” explores something that all of the major encyclopedia publishers have talked about for several years: how to develop an “intelligent encyclopedia” that uses natural means of accessing and using knowledge. Included with this article is contributing editor Mark Dahmke’s look at “An Ideal Video Peripheral,” a glimpse at how personal computers and videodiscs might communicate more efficiently.

In ‘”televisions as Monitors,” Ken Coach describes some of the characteristics common to the new generation of television receivers that can double as microcomputer video monitors.

If you already have an inexpensive VCR with limited or no programming capabilities or are considering purchasing one. Cy Tymony’s “Computer Control of a Video Recorder” should be particulary interesting. This construction project enables you to use your micro as a programmable control center for your VCR.

Stan Jarvis’s “Videodiscs and Computers” takes a look at the videodisc industry, its evolution. and the myriad of companies and equipment facing personal computer owners.

As an example of what you can do with a CAV (frame-addressable) videodisc, Rod Daynes and Steve Holder designed a game around a generic version in “Controlling Videodiscs with Micros.” They used the videodisc support commands available in the Sony SMC-70 computer.

—Gene Smarte. Managing Editor

ELECTRONIC ENCYCLOPEDIAS

by Peter R. Cook

Interactive video technologies help explore “the realm of worthwhile knowledge”.

Peter R. Cook is vice president of creative services for Grolier Electronic Publishing Inc. (95 Madison Ave., New York, NY 10016). He is responsible for coordinating the development of that firm’s Multi-Component Electronic Encyclopedia, which is intended to combine videodisc and videotex technology. Cook spent five years at Arete Publishing Co.. producing the Academic American Encyclopedia.

ENCYCLOPEDIAS EVOKE special memories for most people. When you were a child, you probably did research for essays and class projects, looked for “naughty” words and pictures, struggled through dull facts about dead people, or discovered fascinating facts in articles from “Aardvark” to “Zygote.” Later on, maybe you used an encyclopedia to solve crossword puzzles, to help a child with homework, or to look up the location of a vacation spot or an unfamiliar country mentioned in the news.

Now there is a whole new range of associations that most people wouldn’t normally connect with an encyclopedia: on-line databases, videotex systems, and laser videodiscs—new interactive technologies that, at first glance, might appear to be the very antithesis of the traditional printed encyclopedia. But major electronic publishing companies are beginning to create a new generation of encyclopedias, powerful informational/educational tools that can interact through, and with, all of these new media.

As part of a long-range development plan, Grolier is creating a massive encyclopedic database of text and audiovisual materials that can be accessed and manipulated using interactive technologies. The two major components of this plan, the text database and the audiovisual database, currently are be- ing developed along separate but convergent tracks—one utilizes on-line or videotex technology, the other utilizes videodisc technology.

In this article, 1 will review both development tracks, observing from the publisher’s point of view how information is enhanced by delivery via new interactive technologies. But let’s begin with a look at encyclopedias delivered via a traditional interactive medium—the book. Though new technologies should not slavishly imitate those that preceded them, there is much to be gained by using the book metaphor for building the foundation of an information tool that utilizes the full power of interactive video technologies.

According to an entry in the Academic American Encyclopedia (AAE), a printed general encyclopedia “attempts to present the entire realm of worthwhile knowledge: the humanities and literature: fine, applied, and performing arts: science and technology: history and social sciences: critical issues such as bioethics and civil rights: and select data on significant places and persons.”

Mechanics and economics tend to limit the size of a printed encyclopedia’s “realm of worthwhile knowledge”: of necessity, the information itself is synthesized and summarized. Distillation and outlining of knowledge, along with broad coverage, make a general encyclopedia—whether it be in 1, 20, or 30 volumes—a useful reference tool.

The reference characteristics of encyclopedias are what brought about their rather unnatural, usually alphabetical, information structure. There is no inherent logic in grouping “Aardvark” with “Alvar Aalto” and “Hank Aaron,” although such curious juxtapositions often create wonderfully serendipitous discoveries. Diverse subjects are thrown together for no reason other than the fact that an alphabetical, dictionary-like organization improves access to information. Some encyclopedias still cling to the earlier thematic approach in which related information is grouped together. Thematic groupings make it possible for information to be viewed in a broader context, but highly specific facts are harder to locate.

Alphabetical encyclopedias compensate for the seemingly arbitrary arrangement of their articles by using in- tegrated cross-references that indicate the presence of related articles and draw connections between the various realms of knowledge. In its current (fifteenth) edition, the Encyclopaedia Britannica attempts to overcome the shortcomings of this alphabetic tyranny by breaking the set into three resources: Propaedia. a one-volume overview of knowledge; Macropaedia, a collection of in-depth articles (many of considerable length) on broad areas of knowledge; and Micropaedia. containing very short articles as well as discrete, specific facts. The attempt, however, is thwarted by the awkwardness of three integrated, but separate, resources.

While encyclopedias may have to struggle with the drawbacks of their own unique organizations, they also exhibit the positive characteristics of the printed book. Books are physical entities, portable random-access devices. Their organization is universally recognized: pages, chapters, tables of contents, prefaces, introductions, indexes, bibliographies, etc. They are the framework of written language. Other characteristics inherent in books let you browse through them easily and give you a sense of scale and place. You know how to navigate in a book.

But the book also has its limitations. It is a fixed medium: once printed, it cannot be changed. The only way for a user to update a printed encyclopedia is to buy a new one, write notes in the margin, or purchase the yearbook that most encyclopedia publishers issue annually to maintain the currency of existing sets. The information is also fixed in that it cannot be dynamically rearranged for the user’s convenience. Comparing all the articles on dinosaurs, for example, might require accessing more than a dozen volumes. Information access is also limited by the specificity of the article titles and by the quality of the index—nonindexed information is almost the equivalent of no information. Finally, no matter how descriptive the text or how informative the illustrations, no printed encyclopedia can capture the power of a place, person, or event more vividly than an audiovisual medium. Yet the printed encyclopedia is, and will continue to be, a highly valued information resource for most people.

Now consider an electronic encyclopedia that uses the full power of new interactive media and is designed to meet the needs of a new generation of users.

Electronic Reference Work.

In 1982, Grolier acquired the rights to the Academic American Encyclopedia, a new 20-volume general-reference work designed for use in homes and schools. Containing approximately 30,000 articles and 9 million words, the AAE is characterized by its currency and its short entry format (its articles have an average length of approximately 300 words). For us it had the additional virtue of being typeset entirely with computerized equipment, and thus it could be converted for on-line dissemination.

The electronic edition of the AAE has been in existence for two years and is currently available to over 250,000 online and videotex subscribers through existing information utilities. These include services such as CompuServe, Dow Jones News/Retrieval, BRS. Dialog, and Vu/Ttext. We also distribute through NAPLPS (North American Presenta-tion-Level-Protocol Syntax, or “nap-lips”) graphics-based systems: Viewtron, Keycom, and Times Mirror’s Gateway (which uses Telidon graphics, a forerunner of NAPLPS). Users accessing the encyclopedia remain customers of the information utility, which in turn pays Grolier a royalty.

The encyclopedia’s inherent data structure has to be adapted to the display characteristics and access protocols of individual systems. Because these system requirements have a considerable effect on how the user interacts with the encyclopedia, it is worth reviewing some samples at length, beginning with the less complex, and consequently less powerful, systems.

Viewtron: An ASCII/NAPLPS Hybrid.

Viewtron, operated by Viewdata Corp. of America (a Knight-Ridder subsidiary), is the first commercial videotex service to use NAPLPS color graphics, which can be accessed only by AT&T’s Sceptre terminal. This regional service is now available in just three Florida counties; if successful, however, the service will become available in major cities around the country.

Viewtron is a relatively simple menu-driven system that stores most of its databases on preexisting frames. However, because the AAE is a large database (by videotex standards), it is actually accessed through a communications gateway. The AAE text is housed on computers at Vu/Text (another Knight-Ridder subsidiary) in Philadelphia; the computers are linked by dedicated line to the Viewtron host in Miami. A user accessing the system is connected via the gateway as soon as he selects the AAE from a menu (see photos la through Id). The videotex terminal “paints” the appropriate NAPLPS frame, but it has an active window for displaying the ASCI! (American National Standard Code for Information Interchange) text from the Vu/Text gateway. The user is then prompted to type in a search term, which is matched against the AAE’s 30,000 article titles. If an exact match exists, the system displays the first 15 lines of the article and the user can “page” through the rest at his own pace. If the search term is too broad, a number of articles are selected. For example, if just “Lincoln” is entered for information on Abraham Lincoln, the system locates all articles with the word “Lincoln” in the title, including “Lincoln, Nebraska” and “Lincoln Center for the Performing Arts.” The first article is displayed with the qualifier “1 of n [articles],” and the subscriber can then use the system’s “browse” function to scan the first frame of each article.

ASCII (Non-Graphic) Videotex Services.

Dow [ones News/Retrieval and CompuServe are both consumer-oriented information utilities that primarily use keyword and menu-driven access. A subscriber logging onto Dow Jones News/ Retrieval is guided to the AAE via names and enters a search term or query that is matched against article headings. If the search term is in more than one heading, the system generates a menu, listing all of the articles, and the user selects from this (see figures 1a through 1e). If the selected article is long, an additional menu (or series of menus) is shown, providing a numbered outline of the contents. Thus, the user can select the most appropriate section without having to page through the entire article.

Dow Jones News/Retrieval also lists the number of pages or screens of text in each article and lets the viewer know which page he is viewing (2 of 14, etc.). CompuServe gives each page an individual number, which can be used in conjunction with the GO command to go directly to the page, bypassing the intermediate menu stages. As useful as these features are, menu-driven systems are limited. Access is through the article title only, which doesn’t open up the full potential of an electronic encyclopedia.

Free Text Databases.

Two much more powerful on-line database systems are BRS and Dialog. Considerably more expensive than the consumer databases, BRS and Dialog use free text search procedures in which every word in the database is indexed and therefore can be searched.

When accessing via BRS, a search term or query is entered using the required command language. However, if the search term is not properly qualified, the system may locate far too many occurrences to be helpful. For example, a search for “Einstein” produced several hundred “hits” (occurrences) of the word in 69 articles throughout the encyclopedia (see figures 2a through 2c). But the search can be narrowed down by using Boolean operators such as and, which combines two search terms in the same article: same, which combines two terms in the same paragraph; with. which combines them in the same sentence; and adj. which requires the two terms be directly adjacent. If what I really want to find out about Einstein is when he moved to Princeton, I would phrase the search as: “Einstein same Princeton,” which means I’m looking for the paragraphs in which both of the words appear. This narrows the hits down to just two paragraphs, which can be quickly displayed. These paragraphs tell me that Einstein moved to Princeton in 1933 and that he died there in 1955.

This simple example doesn’t really illustrate the full retrieval capabilities of free text systems. They can be powerful tools in the hands of an experienced researcher or librarian. On the other hand, their complex command syntax and Boolean search logic make them too complex for easy access by most untrained users.

In an effort to broaden the appeal and usage of their services, BRS and Dialog have both introduced easier-to-use and less expensive off-peak services: BRS After Dark and Knowledge Index, respectively. Communications software is also being introduced; Sci-Mate and In-Search simplify command procedures and enable the user to develop a search strategy before going on line. Such efforts are the first steps in what will undoubtedly be a series of software products aimed at making these information utilities (including Dow Jones News/ Retrieval and CompuServe) more powerful and easier to use.

Earlier i used the printed book as an example of a medium with which we are manifestly familiar; the reason for our familiarity is that the structure and conventions of books have been evolving for centuries. On-line databases have been in use for little more than a decade, and it is only in the last two or three years that large numbers of untrained users have started accessing them. Consequently, the learning curve for everyone—system operators, information providers, and end users—is particularly steep. There is still much to be learned about how users interact with electronic information utilities, what kinds of displays are best, what accessing protocols and commands are most effective and easy to learn, and what information is most appropriate.

Enlarging the Database.

The electronic edition of the AAE is already quite different from its printed parent. Updated twice a year instead of annually, it has no physical growth limitations, unlike the printed work. We intend to use this essentially unlimited capacity for growth: to respond directly to users’ needs; to reflect areas of strong current interest; to broaden the database so it is more appropriate to different age and interest levels; and to develop satellite databases designed to interact with the encyclopedia.

Responding to users’ needs: An electronic encyclopedia has a unique advantage over its print counterpart because it is possible to “capture” such key parts of the information transaction as search terms and usage time. Analyzing the captured data can reveal shortcomings, whether in the form of inadequate article headings or missing information. Such data also reveals areas of high and low interest—a useful guide for database growth.

Reflecting areas of strong current interest: Printed encyclopedias contain little information of transitory interest. It will be a long time before the Britannica has an article on Michael Jackson. This isn’t necessarily because encyclopedia editors disdain popular culture. The physical limitations of the printed work make it difficult and costly to insert large numbers of new articles each year. Space has to be found for each new article, usually at the expense of other articles. Further, an article of fleeting interest may create a difficult hole to fill when its importance diminishes.

This is not the case with an electronic edition, in which articles can be added to and deleted from the database with considerable ease. For example, coverage of the Olympic Games and athletes can grow in anticipation of this year’s meeting in Los Angeles. Next year the coverage can be reduced. An electronic encyclopedia can be a truly responsive, dynamic reference work.

Expanding for a broader audience: In its present form, the AAE spans a wide range of age and interest levels. As stated in its preface, the AAE is “for students from junior high school, high school, or college and for the inquisitive adult.” By expanding that base to include young children and advanced scholars, the resulting database will be several times larger than the printed encyclopedia and will be capable of responding appropriately to the user’s age and interests.

Satellite databases: Grolier has recently completed the first of a series of satellite database products designed to interact with, and be enhanced by, the electronic encyclopedia. Whiz-Quiz is a menu-driven educational game that directs the player to the AAE to find out more about a topic. We believe that children in particular will be compelled by this mechanism to explore the encyclopedia.

The User Interface.

Regardless of what shape or direction a database takes, the key to its use and value is the user interface. The user interacts with the electronic encyclopedia on several distinct levels: entry level: logging onto a system and getting to the database search level: entering a search term to locate a specific article (or group of articles) retrieval level: once the relevant article is located, finding and retrieving the required information manipulation level: getting the information in the form of a printout, writing words down directly from the screen, or simply remembering the facts exit level: leaving the database The first and last levels are entirely the province of the system’s operator and, in any event, are not an obstacle to most users. The search and retrieval levels, however, are areas of major concern because that is where the user works most closely with the system. The information-manipulation level will become increasingly important as software is written to take full advantage of encyclopedic databases.

The best way to analyze potential improvements at the search level is to return to the book metaphor. As an access device, a book is very forgiving. When you look something up in an index, you usually don’t need to know the exact spelling to locate it. Likewise, you sometimes use a dictionary because you can’t spell a word but have no problem locating it.

Databases are not so forgiving. A misspelled search term, no matter how close to the correct word, cannot be used to locate the required article. Some systems attempt to overcome this by providing a function called truncation. On Dow Jones News/Retrieval, for example, all you have to enter for an article on Zbigniew Brzezinski is “BRZ—a nice feature, but not the complete solution. The problem isn’t just misspelling; children in particular tend to use plurals for certain common nouns: cats, dogs, trees, dinosaurs, etc. This is not a problem in a printed reference work; however, when entered on a videotex system, the search terms will fail to match the exact article titles, which are singular.

First-time users, especially children, make repeated errors when entering search terms. Analysis of the search terms for one of the videotex services reveals that about one-third of all terms failed to locate an article. In approximately 90 percent of those cases, the information existed but errors (misspellings, use of plurals, incorrect positioning of names) prevented the user from finding it, at least on the first try.

Clearly, the unforgiving nature of search-term entry on videotex systems is a frustrating inadequacy that can be improved by the system operators. At the same time, the information provider has a responsibility. Other aids to access are required. Current videotex systems allow only keyword access to article headings. There is no on-line index, and today’s videotex systems do not have the full text-indexing capabilities of BRS and Dialog. Clearly, the specific entry headings need to be broadened so that the same information is available via several different search terms. An on-line index would broaden access still further, especially when combined with a thesaurus function.

Free text systems are not restricted to keyword access. Rather, the user can focus his search language to a highly specific degree, examining the body of knowledge with precision. The price of that precision is a high degree of practice and skill.

In the long term, both keyword and free text access to large databases probably will give way to search languages with a high degree of artificial intelligence (AI). At present, when you search for information in an encyclopedia, particularly on videotex systems, you have to think about its likely location. You cannot interrogate the database, but this is exactly what you should be able to do, posing such questions as: Who wrote The Grapes of Wrath? How many Nobel Prize winners went to Harvard? Where do icebergs come from?

Answers to all these questions can be found eventually with current systems, but a fully developed database incorporating AI search techniques would take you directly to the sources.

Having located an article, the user begins to read it. The “window” into the encyclopedia’s massive database is a television or monitor. The text display (depending on the service and the end-user’s hardware) varies from 16 lines by 32 characters per line (approximately 85 words) to 2 5 lines by 80 characters per line (approximately 330 words). By contrast, the printed AAE contains 1500 words per page, and pages can be viewed two at a time. Clearly, current video-display technologies are capable of only the most myopic view of a large text database, which is why it is all the more important to be able to rapidly shift the view, to be able to browse and move around in an article quickly and easily.

But there is a need for other orientation tool§, such as sequential numbering of article pages (which Dow Jones News/Retrieval has) as well as individual numbering of pages (which CompuServe has).

A recent study of a group of eighth-graders’ use of the videotex AAE produced some interesting findings. While the students searched for articles and moved around in them with varying degrees of proficiency, they confined their activities to finding information rather than using it. They actually read and manipulated the information later as printouts, which could be studied at leisure, marked up, and incorporated into their research projects. In fact, I suspect that many of our users who have access to a printer do their serious reading in ink-on-paper form.

This doesn’t mean that no one reads text from the screen. Graphics-oriented videotex systems, such as Viewtron, are based on the assumption that subscribers will read from the screen. This is fine for news summaries. But the real utility of an electronic encyclopedia won’t be realized until people can access, manipulate, and reorganize significant amounts of information electronically using such powerful information tools as word processors, database managers, and graphics programs.

Electronic Knowledge Land.

Grolier is working on some of the refinements I have been describing. We believe that, having freed encyclopedia information from the artificial constraints of the alphabet and the physical constraints of the book, we should be able to reshape and reorganize that information. We want to put it to new uses. We want to create a reference tool that can interact with other information tools. Additionally, we’re trying to build a conceptual framework—an outline of knowledge—for intellectual pursuit and stimulation. At present, this new “knowledge land” is largely uncharted, although major landmarks are known. We hope that videotex systems will have the navigational tools to explore it fully, and that software producers will have created the manipulation tools to exploit it fully.

A Videodisc Encyclopedia.

The videodisc is another interactive medium that we believe is applicable to encyclopedia information. The ability to randomly access any frame, combined with the disc’s dense storage capacity (54,000 frames per side on a laser disc) and its inherent audiovisual properties, make it a particularly powerful publishing medium. As part of a joint venture with Long- man, a British publisher, Grolier recently produced a pilot disc. The pilot was developed as a test vehicle to determine how the organization, content, and audiovisual treatment of encyclopedia material might best be accomplished.

Long before the pilot went into production, we had concluded that we would need a lot of discs to encompass all the information contained in a general encyclopedia. This led us to ask ourselves how each disc should be organized so that it could be both a stand-alone information resource and a part of an integrated series. We decided to organize each disc around a specific theme or subject area. The pilot is part of what will eventually be a two-sided disc devoted to the human body (see photos 2a through 2c).

Each thematic videodisc will be a self-contained information resource. The discs will not attempt to be the equivalent of a printed reference work. Rather, each disc will “illuminate” knowledge areas, conveying through audiovisual means only the essence of a subject.

Designed for use with a standard consumer laser videodisc player under normal keypad control, the discs will become considerably more versatile resources under microcomputer control. While the number of combined microcomputer/videodisc applications has increased substantially in the last two years—applications that include training, point of purchase, education, and games—there are few truly “generic” discs for which software can be written.

Grolier is developing two electronic databases, one in text form and one in audiovisual form. These two databases are being developed separately so that each can take advantage of the separately developing markets for on-line / videotex and videodisc products. But both databases will ultimately be brought together (although whether through telecommunications or local mass storage is yet to be determined). The result should be an innovative informational/educational resource: an encyclopedia that is appropriate to the media and appropriate to the times.

An Ideal Video Peripheral by Mark Dahmke

As a software consultant, my major complaint with most of the popular videodisc players is that they communicate with computers very poorly. Typically, the videodisc player is treated as a printer or a plotter; the user has to deal with commands that may or may not be ASCII (American National Standard Code for Information Interchange) format and may or may not be logical and consistent.

The Discovision (now Pioneer) model 7820 had a command set that looked like a selection of random numbers. The codes to send the numerals 0 through 9 were: 3F. 0F, 8F. 4F. 2F. AF. 6F. 1F, 9F, and 5F. in that order. In addition, the 7820 had an IEEE-488 external interface that wasn’t compatible with most microcomputers. In an attempted remedy. Discovision built a serial converter box to change the IEEE-488 protocol to and from RS-232C levels. What they came up with was a protocol converter with a 1-byte buffer that could easily be overrun, erasing the last command before it could get to the player.

Even if the translation and protocol conversion problems are ironed out eventually, a programmer is still faced with a stiff challenge in trying to get the status and frame number back from the player. Some players won’t give out this information at all. Ones that do return strangely encoded bytes that take many instructions to untangle. On some players the frame number comes back as a 2-byte integer. on another it comes back as four ASCII digits in hexadecimal, and on still another it shows up as five-decimal ASCII digits. Any software expected to run on more than one model of videodisc player has to account for all of these differences.

Timing seems to be the worst problem with interfaces. Data sent to the player at serial-port speeds just can’t control a fast videodisc player in the manner required by modern interactive applications. Some new players offer parallel ports, but many computers (e.g.. the IBM Personal Computer) don’t support full bidirectional interfaces. IBM claims that its Centronics port is “parallel input and output.” However, if you check the circuit diagram, you will see that it isn’t. It is wired so that reading the port returns only what was last sent.

Newer videodisc players operate at floppy-disk and, in some cases, hard-disk speeds. Some have worst-case access times of 2 to 3 seconds. Within a year or two, I expect to see write-once, multiple-read videodiscs with interfaces to let them be used as archives. (Some current videodiscs can hold gigabytes of data.) For this to work, however, the interface will have to be smart enough to recover the stored data and fast enough to return it to main memory at magnetic-disk speeds. This technique can work, as shown by the fact that it is already being used in several hard-disk backup systems for videotape recording equipment. In these disk-to-tape systems, the data from the disk is recorded redundantly in the scan lines of a National “television System Committee (NTSC) signal, which is then recorded on videotape. While this prac- tice could easily be transferred to videodisc hardware, much of the videodisc would be wasted and not used to its full potential. The developing direct-digital recording techniques will remedy this problem by maximizing use of the disc recording surface.

Loading software into a personal computer from a videodisc as if it were a floppy disk would greatly enhance educational applications. For example, audiovisual and computer-graphics course material (all orchestrated by an authoring language) could be combined and loaded into a personal computer from the first part of a videodisc while just the audiovisual portion is stored separately on the remainder of the videodisc. A development system would consist of the videodisc player and other end-product hardware, but the graphics and curriculum-specific data, or “courseware.” would be developed on attached floppy-disk or hard-disk systems.

Figure 1 shows one possible hardware configuration for a first-generation intelligent player. The main feature of the design is the videodisc interface adapter, which would plug into an expansion slot on the microcomputer. The interface adapter gives the programmer tight control over the timing of the player and also controls the video overlay circuit.

As digital television and audio reproduction become affordable and popular (I estimate that this will take five to seven years), we’ll be able to define the formats that will let personal computers store and retrieve video images and sound. We’ll be able to create high-resolution computer graphics and synthesized music on personal computers and write it onto a write-once videodisc peripheral. We’ll then be able to play it back through digital television sets. Alternately, digital television images could be recorded from TV sets onto a videodisc and then retrieved. displayed, or processed on personal computers.

Figure 2 shows a second-generation interface built around digital television. As 32-bit processors become faster, and memory bandwidth greater, it will be possible to directly manipulate high-resolution images that come from the videodisc or are created directly by the microcomputer. The video output from the TV camera can be routed to a digital television for viewing, or the output can be held in the graphics frame buffer for further modification and processing.

Mark Dahmke. a contributing editor for BYTE, is a software consultant and heads MCD Consulting Inc. He can be contacted at POB 80266. Lincoln. NE 68501.

view additional pages

What’s a RAM?

The vocabulary of engineers or experimenters working with computers, synthesizers, electronic calculators and similar digital devices is replete with acronyms you should know. RAM is one, read on to find out what it is and how it’s used.

by DON LANCASTER

ANY MEMORY IS A STORAGE DEVICE THAT is given some information at some time and hopefully will return that identical information at a later date for reuse at least once. The most elemental unit of a memory storage system is the cell which can store one bit consisting of a “1-0″ or “Yes-No” simple decision. Memory cells are often grouped into words of several bits each. These words can represent the number in a calculator, an instruction command in a computer, a tone and its duration in an electronic music composer, an alphanumeric character in a TV Typewriter and so on.

Memories can range from one bit to many billions of bits. The equivalent of the human memory is sometimes suggested as 10 billion bits while the longest memory you can buy in a single off-the-shelf integrated circuit is 4096 bits.

There are several different types of memories. You usually classify them by when, how and how often you put information in them. A Read Only Memory (See “What is a Read Only Memory?”, Radio-Electronics, February 1974) has information put into it only once. It keeps the information inside it more or less permanently. Read only memories are often used for such things as square root, log and trig instruction microprograms in a calculator, for time-zone conversion in a digital clock and for many other situations where you always want the same response to your system. Some read-only-memory systems are called “table lookup” systems, for they provide an “answer” in the same way you would get it from a math handbook. A few read-only-memory systems can be altered, but not rapidly. This is done by erasing them with intense ultraviolet light and reprogramming them; others are altered with special voltage or current pulses. Sometimes these are called Read Mostly Memories.

We could also theoretically have a write only memory that would accept information but never return it. Contrary to some misguided and uninformed industry jokes about WOM’s, these DO “exist and have a very specialized use in computer programming, particularly in data stripping and formating. Still, nobody really manufactures WOM’s. When you want a WOM function, you use one location of a read-write memory over and over again instead, never bothering to read it.

The most versatile memory is one that you can write (put information into) and read (receive information from) rapidly and in any sequence. Magnetic cores are typically this type of memory, although by eliminating or not using the write current generators, we can also obtain a read only function. Most cores are destructively read out, meaning that the information is lost the first time you use it. You then have to perform a rewrite after read operation and then put the information back into the memory cells if you are going to use the information again.

Most semiconductor memories are non-destructively read out in that you can accept information without physically altering the memory contents.

If you must put the information in and get it back in one specified sequence, you have a sequential memory. Long MOS shift registers can make a sequential memory. These have traditionally been lower in cost than true read-write memories, but have disadvantages of being noisier and having to wait a long time for the information you need to come out.

The more versatile read-write memory is one that you can read or write in any location at any time. This is called a Random Access Memory, or RAM for short. RAM’s can be made sequential simply by deciding that you want to access or address things in order.

A memory is non-volatile if you can remove the supply power or stop moving the data around inside the memory and still hold the information. Magnetic core is usually non-volatile. Semiconductor read only memories are, of course, non-volatile. Most reasonable or available semiconductor RAM’s are volatile and you must keep the supply power up or you will lose information. Many RAM’s offer a reduced power mode where you can keep information for long times on battery power. In a few years, we can expect true non-volatile semiconductor RAM’s, but for now, you have to design your memory application in such a way that the information is either no longer needed or stored somewhere else than in a semiconductor RAM during power down times. This really isn’t nearly as bad as it seems for usually you can easily get around the problem one way or another. Often a mixture of ROM’s and RAM’s in a single system does the job.

There are two basic types of semiconductor RAM’s. These are the static RAM and the dynamic RAM. Both of these are volatile and will lose information during power down times. The difference is that static RAM’s will keep their information so long as power is applied without reshuffling or refreshing the data while a dynamic RAM has to have its internal storage moved around occasionally, often at a 500-hertz rate or faster. Static RAM’s usually have a flip-flop cell for data storage. Once set or reset, it will stay in that state until power is removed or it is relr tten. Dynamic RAM’s usually use a capacitor for data storage. The capacitor will eventually discharge and thus the data must be moved or refreshed before it is lost. Dynamic RAM’s are normally far cheaper as you can pack a lot more bits onto a given size chip, but they add to the external circuit complexity and may take some elaborate timing to reliably get them to work. Thus, dynamic RAM’s are more suited for very large memory systems, those over 50,000 bits or so.

Today, you can buy a 256-bit surplus static RAM for $2.56 and get the same thing new for under $6.00. A 1024-bit dynamic RAM runs around $5.00 surplus and under $12.00 new. Thus, we are talking prices right now of a penny per bit and under and projected pricing runs as low as one-tenth of a penny per bit. At this projected pricing, a minicomputer computer memory big enough to speak Basic or Fortran could be built for a memory component cost of $64.00 for 64,000 bits, perhaps arranged as 4000 words of 16 bits each. Simpler memory systems for things like terminals, electronic locks, music composers and a whole bunch of things nobody has thought up yet today should cost well under $20.00 and eventually should come down to $2.00. So, now is the time to start becoming familiar with these exciting new devices.

A simple semiconductor RAM Let’s start with a rather small RAM and see what we can do with it. We’ll use the 7474 TTL dual type D flip-flop as shown in Fig. 1. We’ll start with a one-bit memory and then double it to two bits by using both halves of the package.

In Fig. 1-a, we use half the 7474. This stage can store a “1″ (often a high state around 3.3 volts) or a “0″ (usually a low state around 0.5 volt). The stored value appears at the “Q” output. The opposite or compliment of the stored value appears at ‘he Q output. We have a data or D input and a clock or CL input. Information present on ‘he D line gets loaded into our memory at the time the clock goes from ground to a positive value. To enter information into our memory, we put the information on the D line. At that time, it does NOT go into the cell. At the instant we bring the clock line from ground to a positive level or from a TTL positive logic 0 to a positive logic 1, we actually load or write the information into our flip-flop cell. Whatever was on D at the instant of positive edge clocking gets loaded into the memory and appears at the Q output.

This is a random access memory as we always can get to the memory cell (trivial, as we only have one cell) anytime we want. It is static as it will keep the loaded information for as long as we apply power. It is volatile as the information will go away if we ever shut off the +5-volt power supply. And our simple memory is organized as “one word of one bit each.”

We can watch or read our memory any time we like, but since a change may be produced during clocking, we shouldn’t be using or reading at that particular instant. We call the clocking interval the write cycle. Time spent looking at this particular cell’s output is called the read cycle. Normally, you don’t read and write simultaneously. You either execute a read cycle where you monitor and use the output of the memory cell or you execute a write cycle where you place new information into the cell. The 7474 will do a write or a read cycle in under 50 ns. Since nothing physically changes internal to the 7474 during reading, the readout is non-destructive and we can reuse the stored information hundreds or even millions of times if we like.

A one-bit, one-word memory by itself isn’t too useful, although you can think of an alarm system as a one-bit memory and there are numerous other trivial applications. To do any really useful function, we most often need quite a few more bits of storage.

Figure 1-b shows how you can use both halves of a 7474 to build a memory of one word of two bits. This is done by simultaneously clocking each half of the package and using both outputs at once. Thus we have two data input lines, two output lines and one write line. This organization is one word of two bits. In 1-c, we have the opposite, a memory of two words of one bit each. Now something new has been added. We have to combine or select which of the two memory bits is going to appear as an output. We also have to decide which of the two cells is going to have data written into it at any given time. This decision is called addressing. We now have to address cell A (a 0 on the address line) or cell B (a 1 on the address line). By controlling the address line, we select which memory cell is to be acted upon or read.

The more cells we have, the more complicated the addressing will become. Note that we needn’t alternate memory cells if you don’t want to. You can address either cell in any sequence you want. Hence the name random access.

Adding more bits We could use as many 7474′s as we like to build up any memory, but even at surplus prices, the 25( or so per bit and the large supply power and size will eventually get to us. The next step up is to use packages with more than two D flip-flops. Quad and hex latches, the 74175 and 74174 are a good choice. Figure 2 shows some memory circuits using these components.

In Fig. 2-a, we have a 16-cell memory arranged as four words of four bits each. We have four data lines, four output lines and two address lines. These two address lines are binarily decoded (00, 01, 10 and 11) to get at the four possible memory cell locations. We might use this memory to store four BCD numbers as part of a computer or calculator.

In Fig. 2-b, we use eight 74174′s to build a 48-bit memory organized as eight words of six bits each. This time, we have six data input lines, six output lines, and three address lines. The three address lines are decoded (000, 001, 010, 011, 100, 101, 110 and 111) to get at the eight possible locations of six cell groups. Since we can represent a letter, number, space or punctuation with six bits of the standard ASCII code, this memory could be used to store an eight character message.

Which organization?

Suppose we had a 64-cell memory. How could we group the cells to obtain different combinations of bits-per-word and numbers of words? Figure 3 shows some possibilities. While each of these memories is 64 bits total capacity, the organization of each is different.

In Fig. 3-a, we have one word of 64 bits each. We need zero address lines since we are always looking at the same word, but we need 64 input lines and 64 output lines. In Fig. 3-b, we have two words of 32-bits each. We now need one address line to select which half of the memory is to be written into or read from. There are 32 input leads and 32 output leads. The next combination of Fig. 3-c would be four words of 16 bits each. Here we need two address lines binarily decoded to select which quarter of the memory is to be active and there would be 16 input leads and 16 output leads.

You can rapidly run down the other organizations. Figure 3-d gives us eight words of eight bits each. There are three address lines needed that are decoded one-of-eight to pick one-eighth of the memory for use and we have eight input lines and eight output lines. In Fig. 3-e, we have four words of 16 bits each. Four address lines decode one of sixteen and there are four input and output lines. Two words of 32 bits each take two input lines, two output lines and five address lines, the latter decoded one- of -32 as shown in Fig. 3-f. Finally, in Fig. 3-g, we have 64 words of one bit each. There is one input line, one output line and six address lines which are binarily decoded one-of-64 to pick which of the individual memory cells is to be interrogated.

So, we have a wide choice of organizations to any memory. The more the bits, the more the choices. Which do we use?

This depends on you if you are working with a large system and depends on the integrated circuit manufacturer if you are trying to get the job done with only one or two stock integrated circuits. Obviously, you organize the memory to suit the information you are trying to put into it. Four-bit words are common for BCD (binary coded decimal) number storage in calculators. Six-bit words are often used to store ASCII characters. If the full ASCII code, including transparent control commands and lower case and error detection is to be used, we have to up to eight bits per word. Or, we might like to use the remaining two bits to select a color on a color display. We could get one of four with two bits. Minicomputers tend to use 8-, 9-, 12-, 13-, 16-, 17-, 18-, 24- or 25-bit words depending on the manufacturer and the task the computer is aimed at. So, for system’s use, you pick the number of bits needed to do the job.

On the other hand, if you are a integrated circuit manufacturer, you want to have the most reasonable package in your system. The majority of semiconductor memories only have ONE input line and ONE output line and address lines for one-of-N decoding, giving you organizations such as 256 one-bit words 1024 one-bit words, 4096 one-bit words, and so on. Occasionally a smaller memory may have four bits per word, to make working with BCD numbers easier. Other arrangements are rarely used and you usually add packages to pick up the total number of bits you want.

Decoding All organizations in Fig. 3 have binary to one-of-N decoders on the address lines. If this decoder is internally provided in the integrated circuit as it almost always is, we have an internally decoded memory. If we must provide external address decoding as is common with magnetic cores, we need external decoding. External decoding is also needed when you have several memory packages that you are combining for a total storage. In this case, you use output enable or chip select lines to pick which package is to be used. Once selected, each individual package then goes on to provide internal decoding. For instance, with two IC’s we could simply tie their inputs and outputs together and drive the first memory’s chip select as an address line and drive the second memory’s chip select from the compliment of that line. Thus, we pick one-of-two memory IC’s and the chip selects give us a new form of addressing. If we tie four memories together, we use two new address lines, one-of-four decode them and then chip select only one memory at a time. Figure 4 shows how you can expand memories using the chip select system.

Unlike magnetic cores and many older memory systems, the data input and output lines are completely separate with most new semiconductor RAM’s. This eliminates amplifier recovery problems, steering networks, “single port” problems and things like this.

Who makes what?

Figure 5 is a list of my choice of the best There are lots of different ways to classify semiconductor RAM’s. One grouping is based on the process used. Bipolar RAM’s include TI L and ECL logic. MOS versions include P-channel, (metal and silicon gate), N-channel, and CMOS types. In the past, MOS devices have almost always been slower and much cheaper. Some MOS memories are now as fast as TTL and most MOS devices will continue to be cheaper than bipolar for some time to come.

MOS memories are further broken down into static and dynamic versions. Dynamic versions are much cheaper and much harder to use, particularly in experimental or very small system applications.

Let’s take a closer look at some specific IC’s: 7489 The 7489 is a good choice for initial experiments with RAM’s. It is TTL and works off a single 5-volt supply. Organization is 16 words of four bits each as shown in Fig. 7.

There are four data inputs and four data outputs along with four address lines. The address lines are four-line-to-one-of-sixteen decoded internally. Internal circuitry is arranged so that you store and read out the compliment of the input information.

To read this memory, you apply a four-bit address to pick the slot you want to look at and then bring the memory enable line low. For instance, address 0101 selects the fifth group of four cells. Data appears at the output shortly after the address is stable.

To write into the 7489, pick an address, input the compliment of the data you want to store and then briefly bring the write enable low. This loads the memory.

One thing you have to watch very carefully in any semiconductor memory is that the address cannot be changed immediately before, during or immediately after a write command. (The definition of “immediately” varies with the IC—carefully consult the data sheets!) As a memory address changes, certain locations are “flashed” by in the decoding process. It is possible to write, erase or physically move data around if you aren’t careful. ALWAYS PULSE THE WRITE COMMAND ON ANY SEMICONDUCTOR MEMORY. NEVER CHANGE ADDRESSES DURING WRITE PULSING! Put another way, always leave the memory in a disable or a read mode. Don’t put into write mode until after the address is stable. This particular memory cycles in under 50 ns. If you are running any memory very fast, times will occur when the old information or wrong information will be put out until the answers settle down. If this “garbage” time is too great for your application, you can add a latch to the output (perhaps a 74174) to sample the output only during instants when you know the data is good. A very few new memory lC’s include internal latches and eliminate this problem.

By the same token, if you are running fast, the ripple and gate times on address changes may cut into your cycle time significantly. Again, if you are running fast, it pays to either use fully synchronous timing or else latch the addresses to get them all changing at once. Some semiconductor mainframes get around the problem by using emitter coupled logic (ECL) and its very high speeds for addressing.

The 7489 has a few obvious and apparently untapped electronic music applications. For instance, you can use sixteen four-bit words to completely specify one cycle of a music waveform, the attack-sustain-decay envelope of a note or a melodic sequence. These run around $3.50 surplus and under $11.00 new.

Other TTL memories There’s quite a few other TTL memories available, some as long as 1024 bits. The 7481 is a very old design arranged as sixteen words of one bit each. The 74170 is called a 4×4 file, meaning it is a 16-bit memory arranged as four words of four bits each. The 74200 and faster 74S200 are a 256 x 1 memory or 256 words of one bit each. There’s also a bunch of “non-7400″ TTL memories. The Signetics 8225 is a pin-for-pin replacement for the 7489.

1101 The 1101 is a MOS static memory arranged as 256 words on one bit each. It’s shown in Fig. 8. MOS memories are gener- ally much cheaper and often much slower than TTL ones. The 1101 works on +5, -9 supplies and runs quite hot. There is TTL compatibility on inputs, addresses and outputs. There are seven address lines, internally decoded to pick one of the 256 bits. There is one input line and two output lines, a normal one and its compliment.

To read, you make the chip select low and the read/write low after applying the selected address. The output data will be valid within a microsecond or so afterward.

To write, you select your address, wait 300 ns, bring the read/write line high for at least 400 ns and then wait at least 100 ns after the write line goes low before changing addresses. As usual, NEVER change the address during, before or immediately after writing.

The 1101 is widely available and costs as little as $2.56 for probably good surplus units and as little as 50? for questionable surplus units. New cost is under $6.00. One possible application would to be using six of them in a data terminal or programmable calculator to store a 256-word message using the ASCII code.

Improved 1101′s The original 1101′s were rather slow and could take as long as 1.5 ms to read. They are very hot running and the -9 supply is usually a rather wierd thing to have to provide. Improved devices are now available. An 1101A1 cycles in one microsecond maximum. The Signetics 25L01 and the Mostek MK4007-4P are second-generation, pin-identical, versions that cycle in under a microsecond, consume much less supply power and work on standard +5, -12 supplies.

CMOS RAM’s One new type of 1101 replacement is the CD4061, a CMOS device made by RCA. This is a pin-for-pin replacement, but being CMOS, it takes only one supply and draws utterly negligible supply power if you aren’t writing or changing the address. Thus, you can use this with a very small battery for power down storage and still hold the information.You can also run on incredibly lower currents than the 1101 style devices and much faster as well—several hundred nanoseconds. This makes the 4061 ideal for hand-held data equipment and calculators, as well as meter readers and things like this. The only hitch—it’s a new device and still costs $40.00. Maybe next year.

Other CMOS memories include the Motorola 14505 (64 x 1), the Solid State Scientific SCL5554 (256 x 1) and the Inselek A5503 (256 x 1).

1103 The main reason we include the 1103 here is as a warning NOT to try and use it-—unless you have lots of fancy equipment and considerable experience. This is especially true of surplus 1103′s.

The 1103 is a 1024-word x 1-bit dynamic shift register. It is very low in cost. It ranks as the all time most successful single integrated circuit and it toppled “king core” from the computer world. The device trades a very simple and very dense internal circuit for quite a bit in the way of outside support circuitry. This IC needs critically controlled clocks, usually needs an output sense amplifier, and has a complex timing sequence so elaborate that a 30-ns overlap error in the wrong place will cause information dropout. The 1103 is eminently suited for large memories of at least 50,000 bits (this is tiny by mainframe computer standards) or so, where all the critical support circuitry is easily worked with and may be offset by the savings you get by cramming 1000 bits in each package.

The 1103 uses capacitors for internal data storage. The data must be moved around or refreshed at least 500 times per second.

The 1103 is obsolete today. There are some significantly improved devices available today that are much easier to use, but they still are a rather tough design problem if you do not have elaborate equipment and considerable digital know-how. Improved versions include the Intel 1103-A, the Mostek MK4006 and MK.4008, the Electronic Arrays EA1500 1501 and 1502 and the American Microsystems S3103.

4096 bits The big race today is to build an improved 1103-style integrated circuit with 4096 bits. At least one has been announced by TI at a 100-lot price of $26.00, or around 0.6cc per bit. The other manufacturers aren’t taking this sitting down and the race is on. Pinouts have pretty much been standardized and some should be available as you read this. Competitive products include the Electronic Arrays 1504, the Intel 2107A, Standard Microsystems 4412, Texas Instruments 4030, Microsystems International 7112, Mostek 4096, Motorola 6605 and probably a bunch more. Only eight of these integrated circuits are needed to build a decent minicomputer main memory.

+5-volt, single supply MOS memories A number of new, very easy to use and interchangeable MOS memories are now available that use an n-channel static technology. They have no clocks and are entirely and absolutely TTL compatible. There are no clocks or sense amplifiers needed. These include the Signetics 2602, the Intel 2102, Intersil 1M7552, Motorola MCM6602 and the Microsystems International MF 2102. Cost in single quantities is around 2

These integrated circuits cycle in a microsecond and screened 0.5-ms devices are also available. There are ten input address lines, a data in and a data out. Figure 9 shows the pinouts.

To read, you pick your address with the chip select low and the read-write high. The output data appears within a microsecond or so of an address change. To write, apply your input data, select your address, wait 400 ns, bring the write line low for at least 500 ns, send it back high again and wait at least 100 ns before changing the address. Once again, don't change the address immediately before, during or immediately after the write line is active low. The chip select can be used to expand the memory by several IC's. Six of these in parallel are ideal for a data terminal or TV typewriter memory. Prices should drop well under a penny per bit by next year.

What good are semiconductor memories? Calculators, programmable computers, teaching machines, terminals, TV typewriters, electronic games, minicomputers, fullblown computers, electronic music and hundreds of other applications exist now. What can you do with them? Let us know.

HOW MUCH IS ∛2589¹⁶
The Army’s ENIAC can give you the answer in a fraction off a second!

Think that’s a stumper? You should see some of the ENIAC’s problems! Brain twisters that if put to paper would run off this page and feet beyond… addition, subtraction, multiplication, division — square root, cube root, any root. Solved by an incredibly complex system of circuits operating 18,000 electronic tubes and tipping the scales at 30 tons!

The ENIAC is symbolic of many amazing Army devices with a brilliant future for you! The new Regular Army needs men with aptitude for scientific work, and as one of the first trained in the post-war era, you stand to get in on the ground floor of important jobs

YOUR REGULAR ARMY SERVES THE NATION AND MANKIND IN WAR AND PEACE
which have never before existed. You’ll find that an Army career pays off.

The most attractive fields are filling quickly. Get into the swim while the getting’s good! 1½, 2 and 3 year enlistments are open in the Regular Army to ambitious young men 18 to 34 (17 with parents’ consent) who are otherwise qualified. If you enlist for 3 years, you may choose your own branch of the service, of those still open. Get full details at your nearest Army Recruiting Station.
A GOOD JOB FOR YOU

CHOOSE THIS FINE PROFESSION NOW!

This free booklet will tell you how you can become a Computer Programmer

…and how you can train at home for big earnings in the world’s newest, most exciting profession.

50,000 more programmers needed now! 500,000 more will be needed in a few years!

If you’re dissatisfied with your present job, why don’t you become a programmer? So great is the demand for programmers, you’ll have your choice of openings, with a growing future ahead.

For the vast majority of good positions, a high school education, a logical mind and the right preparation are all that are required. LaSalle, world leader in home education, will train you as a programmer in your spare time.

Why not plan for bigger success in the field where employers are looking for you? Authorities agree that for many years to come there will be more openings in computer programming than people to fill them! Even beginners are paid well; and with experience, programmers can fill major supervisory positions at salaries of $12,000 to $15,000 a year – often as much as $20,000.

Without leaving your present position, you can start your training now — at very low cost. You can study at home, in hours of your own choosing. Experienced LaSalle instructors will guide you by mail every step of the way. Industry experts say this is the most thorough course of its kind that they have seen. The ease and rapidity with which you learn will surprise you.

LaSalle has just published a fascinating booklet “How To Become a Computer Programmer.” In simple, non-technical language this booklet tells you what computer programming is; what the programmer does; how you can train by mail through the school that has enrolled more than 1,500,000 people in its many success courses. Mail the coupon for your free copy or write to LaSalle, 417 S. Dearborn Street, Chicago, Illinois 60605.

LASALLE EXTENSION UNIVERSITY
a Correspondence Institution

417 S. Dearborn Street, Dept. 86-070, Chicago, Illinois 60605
Please mail to me the free illustrated booklet “How To Become a Computer Programmer.”

Human Memory vs. Electronic Brains

Although the complex modern electronic computers are commonly referred to as electronic “brains/7 scientists are not yet able to duplicate the human brain or memory. By comparison, man-made memories are dead and unexciting, according to Dr. F. Joachim Weyl of the Office of Naval Research.

Computer memories and such “brains” as airport traffic-control devices are what might be called set memories, Dr. Weyl explained. The totality of all information that could ever be stored in them is fixed and fully known. It is important, moreover, he said, that complete certainty prevail at all times as to what part of this totality is stored where. Whimsical changes of the information or transfers from one place to another degrade the usefulness of such memories.

Libraries or filing systems are what Dr. Weyl calls “open-ended memories.” The insistence on complete certainty as to the nature and place of everything stored continues, but we do not count on knowing all of the information that is to be stored.

Contrasted to these “dead and unexciting” memories are the memories developed by nature. These are the memories that grow. By performing their function they become bigger and better.

Two such living, growing memories were cited by Dr. Weyl. They are the gene, mechanism of heredity, and the brain.

The gene, he says, represents the memory of the race. It grows in two ways. Sudden random changes in the stored information, called mutations, produce new patterns. And then those new patterns which are better for survival than the old ones are selected in various, quite slowly working ways. This mechanism is not perfect, but it is a self-improving memory.

Neither is the brain a perfect memory mechanism. It experiences some of the difficulties of the librarian in retrieving at the proper moment the material that is stored. But, like the gene, it is self-improving. The use of your brain makes it a better instrument for storing memories.

— Science Service

Sitting here on my 50 mbs internet connection I’m going to say that guess was a bit off. The total amount data they are talking about transmitting over a year is less than the size of the images in this post.

I also particularly liked that the searches on the third page are for “Computer, Privacy Surveillance, NSA and Tapping”. Just a hunch but I’d guess that the person who made that screenshot probably later joined the EFF.

view additional pages

Trends in Telecommunications

On-line search software and faster modems for PCs

by John Markoff

Now that the personal computer (PC) has won the battle for office desktop space, software developers are turning their attention toward programs that combine the storage capacity of mainframe computers with the local processing power of PCs. Although mainframes offer PC users access to huge on-line databases of specialized information, how to get to the information and bring it to the PC in a usable form is another question entirely.

In recent months, a new class of PC software has emerged that facilitates the redistribution of tasks between mainframes and PCs. It is called “on-line search” or “database-access” software, and these programs give us a glimpse of how radically PCs will alter the traditional mainframe database-access model based on one central processor and hundreds of remote dumb terminals.

In contrast, on-line search software uses the processing power of the PC to mediate between the researcher and the mainframe database and can offer potentially both a simpler user interface for novices and a more powerful searching tool for experts.

During the past decade there has been an explosion of new sources of electronic information. Several mainframe electronic-information providers such as The Source, CompuServe, and Newsnet have designed their systems specifically for novice users, but most on-line database services require special training to be used effectively. These include databases such as Dialog Information Retrieval Service, Nexis and Lexis, and Data Resources Incorporated.

The high cost of on-line information is also a deterrent to new users. Some databases on Dialog cost more than $100 an hour. This has meant that users generally must undergo extensive training to learn how to develop search strategies to minimize connect time.

Reducing Costs PC-based on-line search software will be beneficial to database users because it will simplify complex user interfaces now found on many mainframe databases and it will permit extensive off-line preprocessing of searches, there- by reducing the cost of information retrieval.

On-line search software introduced to date can be placed in two distinct categories. The first category is composed of programs that are “loosely coupled” to a specific mainframe database. These programs are extensions of intelligent communications software programs and generally permit automatic log-on, query, and downloading from a host mainframe computer.

The second category includes software that has been “tightly coupled” to one or more particular databases. By tailoring programs for interaction with a host computer, software designers are able to create user interfaces that require little knowledge on the part of the user of either micro-to-mainframe communications or the formal database query process.

The emergence of new communication network standards and standards in the on-line information industry will tighten this coupling to the point where the relationship between the mainframe database system and PC software will approximate the current relationship between operating systems and application programs.

Dialog, a subsidiary of the Lockheed Corporation, is the largest collection of public online databases. It has more than 75 million records of information including articles from over 60,000 journals. These records are contained in more than 200 separate databases ranging form biographic databases such as American Men & Women of Science to statistical databases such as U.S. Exports.

Most Dialog reference records are currently available as abstracts that require you to go to a library to obtain the entire article or source (articles can be ordered on line for an extra fee). However, there is a trend toward making the full text of documents available on line. One Dialog database provider, Information Access Corporation, recently introduced two such databases, called Magazine ASAP and Trade & Industry ASAP, that will cover 120 different popular magazines and publications ranging from Scientific American to Playboy.

In-Search.

In-Search is an example of an on-line search program that has been tightly coupled with the Dialog databases.

In-Search, initially designed to be used on the IBM PC or PC XT, was introduced recently by the Menlo Corporation of Santa Clara, California. This program costs $399. It differs from other database-communication programs both in its scope (the program and its assorted reference files occupy more than one megabyte of disk space) and in the sophistication of its user interface, which offers a window-based display environment and “unhooks” the control of the database query process from the Dialog mainframe computer. “Unhooking” means that you’re able to prepare your query in a screen-oriented editor while either on or off line. The process takes place with little interaction with the Dialog mainframe computer.

If the query has been prepared off line, you can log on to Dialog and have the query sent automatically. When Dialog responds with abstracts, they appear specially formatted in an overlapping window display.

Here, again, the user interaction is not dependent on the control of a remote mainframe computer. If you wish to interrupt the flow of information from the Dialog mainframe, you can do so simply by paging backward or forward through the information in much the same way that you can scroll through a text document in word-processing software. Because information from Dialog can be captured in a buffer (In-Search tailors the size of the buffer to the available memory level of an individual computer) on your PC, it’s possible to selectively mark records for later printing. You also can store retrievals to disk as ASCII (American National Standard Code for Information Interchange) text that can be edited by a word processor or called up for viewing by In-Search.

In designing the In-Search user interface, Menlo Corporation has attempted to take concepts from other popular types of PC software. For example, when working in the query editor, you can edit and change lines of text exactly as though you were working with a document text editor. In-Search has even supplied users with the option of the familiar WordStar command-key sequences for cursor control and word and character deletions (the cursor-control keypad is functional as well). The basic In-Search display also contains a menu of command options that are arranged similarly to those provided by electronic spreadsheet programs. By pressing a function key, you can enter a command mode and select a command that will cause In-Search either to send a particular command to the Dialog system or to retrieve information from its own local database.

Although it is possible to first prepare a particular search strategy off line and then retrieve references quickly to minimize connect-time charges, In-Search is based on a different, more interactive philosophy of on-line database use.

Menlo’s president, Lloyd Kreuzer, argues that In-Search is designed to function in a highly interactive manner. This sets it apart from other PC front-end software packages that assume you know what you want before going on line.

In contrast, Kreuzer believes that the most effective way to use a database like Dialog is to be able to alter a search strategy depending on the nature of the data revealed on a search. “Interactive searching is less precise and therefore more likely to turn up things,” he says. “The keyboard is never dead and [in fact] it is uncoupled from the Dialog process.”

When using In-Search on a fixed disk, the program provides local on-line detailed information on each individual database. This information, traditionally provided as printed textual documentation by Dialog (on forms called “blue sheets”), allows you to obtain information on the scope of an individual database as well as information on specific database indexes that aid in refining searches.

In-Search also supplies you with local context-sensitive on-line help both for using Dialog and In-Search. If you have an IBM PC without a fixed disk, you must insert one of four separate floppy disks that represent major database categories: arts, education, and social sciences; biology and medicine; busi- ness, government, and news: and engineering, mathematics, and physical science. On a fixed-disk PC, these files are directly accessible by the program and in the future it may be possible for the Menlo Corporation to use Dialog to download updates both to the on-line reference sheets and to the In-Search program itself.

The search process begins with selection of a database to search in. The first In-Search display shows three windows. Two small windows on the left side of the screen allow you to select one of the four major categories and to select further specific subject areas within each category. After you select category and subject you can select a specific database. At this point you are placed in the query editor (In-Search calls this the Search Keywords and Phrases screen) to formalize a search.

After In-Search sends the query to Dialog, the references yielded by each search are displayed in a separate window referred to as a reference text display. Any search words that you entered in the query editor appear as highlighted text as they are scrolled on the reference text screen.

At the same time, information on modem status is given in a small window in the lower-right corner of the display. When Dialog is sending records, the window indicates Phone-Working. The status changes to Phone-Online after the records have been retrieved or when you interrupt the retrieval process.

For a simple search to answer the question “Are there any books currently available that describe bicycle tours of the California wine-growing region?” You would first select the Books in Print database and then enter the words “bicycle.” “wine,” and “California” in the query editor. You enter each of these words on a separate line. The first three lines of the editor are labeled S1, S2, and S3. On line S4, you enter the phrase “S1 AND S2 AND S3″ to insure that any reference in Books in Print that contains the first three words in its abstract will be located. (Running this query with In-Search located one book: Grape Expeditions: Bicycle Tours of the California Wine Country.) In-Search documents the AND, OR, and NOT logical operators, which are subsets of Dialog’s complete range. However, expert users can implement all the other search operators that Dialog permits.

Effective searching of the Dialog database, even with In-Search, is frequently complicated. Since Dialog is generally a collection of document abstracts, it is heavily indexed, and it is important to understand the structure of the indexes to conduct a complete search.

A Dialog database is broken into records that are composed of fields. A typical record might include fields such as title, author, journal, abstract, descriptors, and identifiers. (Descriptors and identifiers are standard and nonstandard terms used by the database publisher to identify the subject matter of a record.) Each field is indexed either as a word index or as a phrase index.

In-Search gives you on-line access to specific indexes for each database. You can select an index for any term or phrase entered on the query editor screen. You also can send the Dialog database an “Expand” command that shows a listing of indexed words around the particular search word for a particular field in the database. This often will aid in narrowing down the focus of a search. (It is possible to search only one Dialog database at a time, however, some preselection is possible by searching the subject index first with a special command.) The importance of indexed searching was exemplified when I searched for my last name in The Computer Database. No references were found; however, when the author index was specified, Dialog located 106 references.

Possibly the most intriguing aspect of this new class of software is the change that it portends in the realm of microcomputer-to-mainframe communications. The analogy that casts the mainframe database in the role of an operating system, linked simultaneously to many remote application programs, brings many possibilities into view. In this model, interaction between microcomputers and mainframe computers would be similar to program calls to an operating system.

Menlo’s Kreuzer has called upon online database providers to develop an open-architecture, machine-to-machine interface standard that would permit third-party software developers to create a new generation of applications programs.

“(What is needed is) a universal set of calls to create an open standard for the on-line community that will let us, or anyone, write applications programs,” he says. “The information industry literally will explode once we have a machine interface to all the data.”

Such an architecture would move in a philosophically different direction than the one currently being followed by some on-line information providers who have been setting up systems based on hierarchical “user-friendly” menus for novice users.

Instead, Kreuzer is aiming at fundamentally changing the division of labor between mainframes and PCs. While it is logical that the data searching and sorting algorithms will remain on the mainframe computer, the PC can be expected to handle the user interface, on-line help, and preprocessing of the search request more efficiently.

Further Benefits

Tighter coupling of the communications process between mainframe host and remote PC potentially can yield other dividends as well. Higher data-communications speeds is one obvious possibility. In-Search already uses a significant amount of data compression on the large on-line reference files that are stored on the PC to reduce their size by almost 40 percent. There are a series of simple strategies for increasing the data-communications bandwidth as well. If the applications program can be coupled more tightly to the host computer, it is possible to employ a variety of data-compression strategies to go beyond the current 1200-bps (bit-per-second) limitation over phone lines.

Post-processing is another significant area. While In-Search currently formats only downloaded information and stores it to disk or outputs it to a printer, several other on-line search programs permit later manipulation of information as well. SciMate, from the Institute for Scientific Information (ISI) in Philadelphia, is an on-line search program that is priced at $880 and designed for IBM PC, Apple II, and CP/M computers. It provides for automatic logon and query of four different database systems and includes a local database manager that makes it possible to store downloaded information. The database component of SciMate is called Personal Data Manager. It will take advantage of the record and field structure of information from a host computer or permit you to create your own structure for a local database. Although there are limitations on field and record size, Personal Data Manager permits you to link records to store longer textual documents. You also can move files to word-processing programs or merge locally created notes or documents into the database.

In a smaller fashion. Informatics General Corporation and VisiCorp have developed two complementary programs, Answer/DB and VisiAnswer, that permit the transfer of quantitative data from a corporate mainframe computer to an IBM PC where it can then be loaded into a VisiCalc spreadsheet program for local analysis.

Faster Modems for PCs

Today there are a series of barriers confronting high-speed PC data communications. Most of these barriers fall within the realm of the voice-grade telecommunication network and into existing modem technologies designed to send data over this network.

Yet, while digital technologies are promising dramatically higher communication speeds, a series of new modem designs is being introduced that will bring PC-to-PC data rates up to 9600 bps and, with additional data compression, may push speeds higher.

The new technology wasn’t originally developed for personal computer users, but rather for digital-facsimile transmission systems. Now that the technology has been moved to PCs, it raises a number of possibilities, including using facsimile machines as remote input-output devices for PCs.

Gamma “technology, a Palo Alto, California, data-communications corporation, recently introduced the FAXT-96, a half-duplex 9600-bps synchronous modem board for the IBM PC and PC XT.

Priced initially at $1995 and designed to be used with a synchronous adapter card, the FAXT-96 plugs directly into a card slot in an IBM PC or PC compatible and permits 9600-bps communication over ordinary dial-up telephone lines. The modem includes auto-dial, autoanswer, and multiple-speed features. It connects directly to a phone line and to a synchronous adapter.

The use of dial-up 9600-bps communications is new. It has been made possible because of improvements in modem technology and improvements in the method of encoding digital data on bandwidth-limited voice-grade lines. Control of the FAXT-96 is handled in software from a “master control panel” screen on the IBM PC.

Previous high-speed synchronous modems have been stand-alone units that have been intended for either remote-terminal or micro-to-mainframe communication. The Gamma Technology modem differs in that, although it can be used as a high-speed micro- to-mainframe communications link, a software package also is being offered that permits error-checked PC-to-PC file transmission at 9600 bps.

The shift from asynchronous to synchronous transmission protocols at higher data rates frees the communication process from the start-stop bit overhead, a difference that automatically yields about a 20 percent increase in transmission efficiency.

The significance of higher data communications rates has grown with the deregulation of the communications industry because communication costs are expected to rise. Gamma Technology is claiming that an eightfold increase in data rate (from 1200 bps to 9600 bps) will save several thousand dollars a year if 160K bytes of information are transmitted daily across the United States. Savings would be even greater if data were transmitted overseas.

The FAXT-96 can be programmed to meet several international modem standards set by the CCITT (International Consultant Committee for “telegraph and “telephone). The standards include CCITT V.29 at 9600, 7200, and 4800 bps and CCITT V.27 at 4800 and 2400 bps. Until now, U.S. modem signaling standards have been dominated by AT&T-developed standards. That’s changing, both because of the global need for communications and because AT&T has less influence in an area of deregulation.

There are some limitations. Because sending data at 9600 bps is pressing to the limit what currently is possible with voice-grade lines, poor line quality can make it impossible to send data at that speed. “lb cope with line-quality problems, the Gamma “technology modem automatically tests line quality during an initial handshaking phase and then sets transmission speed at the highest data rate the line will support, ranging from 9600 bps down to 2400 bps. The line test is done by having one system send a known signal to the receiving system. The receiving system knows what it is supposed to get and can make adjustments to make the closest fit.

A recent study by Xerox of facsimile-system performance showed that the same modem technology that Gamma is using would support 9600-bps data transmission worldwide approximately 75 percent of the time over voice-grade lines. Over domestic long-distance lines the 7200-bps rate had to be selected only 27 percent of the time.

In addition to cutting communications costs, higher-speed data communications opens up new applications. Facsimile-to-PC connections would make the transmission of the textual information possible for bit-mapped display on the IBM PC. At 9600 bps the facsimile-transmission time for an by 11-inch piece of paper is 30 seconds. Another possibility is for the transmission of specialized database information that includes diagrams or other images.

With that membrane keyboard, I’d say so.

The Only Video Game Voice Module with an Unlimited Vocabulary

The Voice from Odyssey 2 adds a whole new dimension to the fun of video games. With this optional module and its specially programmed cartridges, Odyssey 2 becomes the only video game system that can repeat any words typed into the keyboard, and much more!

Depending upon which cartridge you insert The Voice can do a whole bunch of other exciting things. Like asking questions and demanding answers to math and spelling problems. It even enhances sound effects and warns of approaching enemies in certain arcade games!

A whole series of specialized arcade, educational and strategy voice cartridges is available for use with The Voice, with a lot more to come. But you can still play all other Odyssey video games through The Voice module.

So if you’re on the lookout for greater challenge, listen to The Voice, from Odyssey 2 The fun you will have will speak for itself. For your nearest dealer call (800) 447-2882. In Illinois call (800) 322-4400.

Odyssey 2
The keyboard is the key to greater challenge.

“I’ve already experienced what happens when one is less than enthusiastic about Macintosh: the Mactribesmen descend in force with fire and sword.”

view additional pages

COMPUTING AT CHAOS MANOR

The AT&T Computers

Jerry Pournelle holds a doctorate in psychology and is a science-fiction writer who also earns a comfortable living writing about computers present and future.

I’ve just come back from COMDEX Winter in the Los Angeles Exposition Center, where I got to play with the new AT&T computers.

Like, Wow!

When AT&T announced a computer line, there was a bit of panic on Wall Street; after the prices were announced, the excitement died away. Too expensive. Who’s worried about a computer line whose lowest-cost item is a $9950 desktop? How can that affect the micro world? That’s what many Wall Street analysts said, anyway.

Dream on.

I don’t own any computer stock—the conflict of interest is obvious—but if 1 did, I’d give that analysis a lot of thought. People, that AT&T desktop computer is one hell of a machine.

True: just now, as I saw it, with little application software, it’s not much of a threat to anyone; but give it time, and there’ll be a different story. The AT&T 3B2/300 desktop provides an awful lot of bang for the buck. If I seriously wanted into the software business, I’d buy a 3B2 development system and get to work polishing up my skills in the C programming language; and I’d expect to get rich from it.

AT&T quotes a four-week delivery time on the 3B2/300. With half a megabyte of memory (plus 32K bytes of ROM [read-only memory] for bootstrapping and housekeeping), it costs $9950 “with standard industry discounts available.” What you get is the basic machine built around the WE 32000 microprocessor chip, one 5/14-inch floppy disk, a Quantum 10-megabyte hard disk, two RS-232C ports, the UNIX System V operating system extended with a number of popular UNIX utilities such as YALOE and YACC (Yet Another Line-Oriented Editor and Yet Another Compiler-Compiler), and the Vi editor from Berkeley UNIX.

The WE 32000 chip is a true 32-bit microprocessor; it corresponds somewhat to the National Semiconductor 32032 or the Motorola 68020. It’s also capable of 8- and 16-bit operations. Moreover, AT&T is committed to heavy-duty silicon support of the system: at the moment, the WE 32000 system has an external memory-management unit (MMU) and other external support chips. The whole system is designed to interface with UNIX. AT&T plans in the future to combine the microprocessor, MMU, and other support into a single chip, thus increasing both speed and reliability; AT&T’s people say the new chips will be upward compatible with the present 32000 systems.

As I write this, memory for the 3B2 is $2400 per megabyte. AT&T makes 256K-bit memory chips and bundles them in clusters to make 1-megabyte boards no larger than a paperback book. Alas, I didn’t think to photograph one of the boards next to one of my paperbacks to illustrate the point, but a megabyte contains about 166,000 English words: the AT&T memory board could hold all the words in one of my novels and have room to spare.

You’d undoubtedly want a full megabyte of memory for a development system, and indeed, to handle full UNIX you’ll probably need two; so the base price for a real-world development system is more like $14,500. We’d also want to upgrade the disk to at least 40 megabytes: that’s $2000 more. Finally, AT&T will sell you a wonderful terminal with a bit-mapped screen and its own WE 32000 processor to run it; that’s an additional $5000, so our development system costs $22,000; hardly cheap.

However: we now have all we need. For an additional $500 per user (less if we want to use less expensive terminals), we can add up to 15 more users. The system is full multitasking, multiuser UNIX: we can let each of our terminals run more than one job. I saw 20 simultaneous jobs running on the 3B2/300, and that didn’t seem to slow it down much.

The Death Star Connection

One of the strangest sights I ever saw was the AT&T “Death Star” logo on an IBM PC; but there it was. You can use the PC as a terminal for the 3B2/300. AT&T, in conjunction with Locus, a small but highly competent Santa Monica software house, has worked out the hardware and software to network the 3B2 with the PC. What we saw running at COMDEX used a 3-Com Ethernet board in the PC and the regular 3B2 Ethernet board; but the software will work just as well with Omninet or (at lower speed, of course) with a straight RS-232C connection. That means we could connect up a Corvus Concept or any other machine that supports Omninet; or, if we want to write the interface driver, any machine with an RS-232C interface.

This gives the 3B2 a lot of flexibility. Even with all terminals and the networking and throwing in a good printer, we’re under $30,000 for a 10-user system with full UNIX.

There’s more, though.

Opening Windows

What I saw running on that 3B2 wasn’t any ordinary UNIX: this UNIX had windows and a mouse, and while it didn’t yet have icons, it had nearly all the capabilities you expect from Apple’s Lisa, only this was loaded down with multiple users and was still fast. Yet more: using that marvelous bit-mapped screen you can run programs under the debugger and see the program output in one window, the debugger in another, and the source code in a third—while having one or two other application programs running along in still more windows. The debugger lets you insert breakpoints, step the program along line by line through the source code, and get the contents of the registers; i.e., it offers all the features you’d expect from a good debugging utility. I’m told it’s an outgrowth of the BLIT windowing UNIX prototype, but it has a lot more capability.

The debugging system takes advantage of the speed and power of the WE 32000 in the terminal. Windowing is managed about the way you’d expect, with pull-down menus and the ability to change window size and shape and shuffle the windows to put the one you like on top; all this is done with the mouse in much the same way that Wirth’s Modula-2 operating system works. I suppose I shouldn’t be surprised, since most of these concepts have previously appeared at the Xerox Palo Alto Research Center (PARC); the whole micro community owes Xerox a vote of thanks for acting as the software R&D center for the industry. I can’t imagine why Xerox doesn’t market some of the goodies that have been developed, instead of exporting them to competitors, but these discoveries sure helped us all.

In any event, if there’s something more powerful than the 3B2/300 for software development at anything like the cost, I haven’t seen it.

The 3B2/300 weighs about 30 pounds and is roughly the same size and shape as an IBM PC XT. Its innards are certainly no more complex than an XT’s. If the introductory quantity one price is $9950, what will the quantity one hundred price be a year from now? Ditto with memory: that $2400/megabyte is quantity one today; I expect it to fall by half within two years.

Do They Know the territory?

Every publication from computer magazines to the Wall Street journal has speculated that AT&T knows a lot about computers but not a lot about marketing. How could The Phone Company learn that? This is, after all, the hated Ma Bell .. .

So, naturally, when I got a chance to interview John Scanlon, AT&T’s vice-president of the Technology Division, even though it wasn’t very original of me, I figured I’d get that one out of the way. “Great R&D,” said I, “but can AT&T handle marketing?”

“We seem to have done a good job of creating a demand for UNIX,” he said.

Which is true enough. Moreover, I notice that even IBM is supporting UNIX, both directly and through Intel, which is developing UNIX for the Intel iAPX286 chip.

“We already have UNIX for our machines. lust turn on the switch: it comes up in UNIX, with most of the features and utilities that programmers want. Who else can offer that at our prices?”

Maybe some others, particularly if you count UNIX clones such as UNOS from Charles River Data Systems, thought 1. but I didn’t want to argue with him.

“And we’re not through. Right now a full UNIX system really needs 2 megabytes, but we’re getting it smaller. We’re 90 percent to getting UNIX tamed. After all, we’ve used UNIX inside Bell for years to provide turnkey systems for our internal use. We do accounting, telephone control, and billing, and we can develop new UNIX shells for each application.”

There’s a lot to that. If some outside company had been supplying Ma Bell with office computers for accounting, billing, word processing, and all the mundane tasks one uses computers for, we’d think it a highly experienced company. AT&T has in fact been using a lot of its own equipment for years; it just didn’t show elsewhere. The company’s got more user-service experience than has usually been realized.

“But,” I asked, “have you really tamed UNIX? It takes a UNIX wizard to keep the system going—at least it does everywhere I know.”

“We have to get the size down,” Scan-Ion told me. “At $4800 for 2 megabytes—and you need that now—UNIX is too expensive for the mass market. But we’ll get it there, and when we do, we’ll get it matched to silicon. Maybe in a few years we’ll have UNIX on a chip.”

Ridiculous, thought I.

Then I remembered where I was.

I was sitting in a carpeted office that looked much. like an expensive executive suite. I’d reached it by going through expensively furnished lounge and reception areas. Of course; AT&T can afford lavish offices—except this wasn’t any lavish office. This was in a bay of the Los Angeles Exposition Center. AT&T not only had the largest exhibit area I’ve ever seen at any computer show: the company had taken an even larger area, closed it to the public, and lavishly furnished it for the convenience of dealers, executives, and the press.

Out in the exhibit area were about 100—and I do not say this for exaggeration—young men and women in dark suits, the most neatly uniformed and well-groomed corps of show people in my memory. Many were extremely polite, with plenty of social skills, but they knew nothing whatever about computers.

If throwing money—or people—at a problem will solve it, AT&T will never have any difficulties. Of course, that doesn’t often work- But then it began to sink in. Along with all the well-scrubbed young women in stockings and heels and pinstripe suits and little neckties, and the young men in similar plumage—in among those were mixed real working software engineers and even a sprinkling of true wizards from Bell Labs. As a matter of fact, it was easy to spot the real hackers: the programmers all read BYTE and knew exactly who I was, while the press-relations people had never heard of me but tried to pretend that they had.

I don’t have to ask if Bell Labs understands transistors. For reasons too complicated to explain, I have framed on my wall a dollar bill signed by Dr. William Shockley: it’s a tiny part of the Nobel prize he received for inventing the transistor while he was at Bell Labs.

The Phone Company never did lack for scientists and engineers. AT&T won’t have problems in that department. This is one of the few companies that plan for decades ahead.

If That’s Not Enough AT&T also has some brilliant management, beginning with Jack Scanlon, who understands this business pretty darned well. We sat in a lavish office suite surrounded by the trappings of corporate power: but Scanlon talks like any true hacker. He reminded me of a lot of the sharper micro people I know: there was that same breadth of knowledge about the field, familiarity with every intimate detail of his own product, obvious pride in the company’s achievements.

He gets particularly excited about what can be done with silicon and the next generation of chips. “Every year we can double the number of transistors we can put on a chip. The guys who design those chips have to do something with all those transistors. We’re seeing a whole new renaissance of silicon architecture. The difference between hardware and software is vanishing . . .”

And more. On software and computer languages: “Language is the wrong way to look at it. Move back a few steps. Watch that guy at his desk. What does he want the machine to do? He knows he can do more than one thing at a time. He has to worry about a lot of things at once. What I want is fundamental building blocks he can throw together fast. This guy wants spelling and maybe a spreadsheet, another fellow needs a database. Once we know the concepts, we don’t have to worry about languages. We can even put them in silicon.

“Obviously, we’re going to support the popular languages, C, and Pascal, and Ada. and Modula-2, but the real goal is to see what people want the machines to do for them.”

That sounds a lot like marketing smarts to me.

Sure: AT&T will make marketing mistakes. The people there are used to figuring out what they think people need and providing it whether it’s wanted or not. It will take considerable effort to modify that attitude.

Consider: here AT&T spent more than I’ve ever seen any company spend at any computer show: this to bring products to a dinky little COMDEX that I wouldn’t have bothered to go across town to see if AT&T hadn’t been there. The press corps was the smallest for any show I’ve been to since the 70s. Having brought the 3B2 to the tiniest pond available, AT&T still didn’t have a good mechanism for getting people to a press conference. The COMDEX staff were even telling people it had been canceled: if I didn’t have a nasty, suspicious nature so that I checked with AT&T myself, I’d have missed it.

just before COMDEX, I received three phone calls, two letters, and a telegram reminding me that I could have an interview with John Scanlon. When I discovered I was missing the press conference—it was in the Bonaventure Hotel, a mile from the show—an AT&T expediting officer put me in a cab to get me there. AT&T sure paid attention to press relations: but with what result? Apple spent far less to get Macintosh on the cover of every magazine in the country. I’ve yet to see real coverage of the AT&T market entry.

On the other hand, AT&T can afford to make mistakes. When I said that at AT&T’s lavish press luncheon, some of the computer press people said, “Yeah, and they got started early.”

Maybe: but that 3B2 is one heck of a machine. It’s state of the art, and it’s available now. Of course, it’s not a retail system. AT&T intends just now to sell to system developers and value adders who’ll package it into full systems that will inevitably cost quite a lot. Retailers needn’t worry just yet.

However: where else can you get a UNIX development system that you can also use for your general-purpose computing at that price? And certainly The Phone Company knows how to build rugged, reliable equipment. 1 wonder what the 3B2/300 will cost next year?

Of course, we haven’t heard the last from the micro world either. AT&T’s machines are still positioned for maximum effect in the minicomputer world, and if AT&T has a year of development time to impact micros, so do first-class microcomputer outfits like Sage and CompuPro. The race isn’t anywhere near decided.

Whatever happens, the micro industry had better think hard about AT&T’s impact, which will be as big as IBM’s. AT&T is big, is here to stay—and has got some damned impressive machinery.

Update: AT&T has also got some real marketing problems. It took me two days to find out the licensing fees for UNIX System V on 68000 systems, and when I finally did, I found the fee structure is set for minicomputers: the company doesn’t know how to handle thousands of sales.

In fact, I found myself conducting a telephone seminar on the micro market. Maybe the people there learned something. More next time.

Big Mac

I’ve been sitting here bringing up Mac-Write, and while swapping disks back and forth on my Macintosh, I thought about the contrast with the AT&T system and UNIX. It takes a couple of minutes to get the Macintosh ready to run a simple text editor: for me that’s far too big a waste of time, especially when the only editor available is MacWrite.

In other words: I’m nowhere near as impressed with the Macintosh as everyone else seems to be.

1 know I’m in trouble for saying that. I’ve already experienced what happens when one is less than enthusiastic about Macintosh: the Mactribesmen descend in force with fire and sword. You must overlook all the Mac’s faults, for after all, they’re only temporary. Everything will be fixed . . .

If IBM or AT&T had come out with a machine that had a single disk drive, no Control or Escape keys, nonstandard interface between keyboard and system, proprietary operating system, limited memory, closed architecture with no possible access to the machine’s innards, disk formats totally incompatible with anything else in the micro community, no languages except Microsoft BASIC (and plenty of bugs in that), and absolutely no application software, the micro community would have screamed bloody murder. Apple has done precisely that, and everyone applauds.

I’m sorry, but I don’t. As I write this, the Macintosh is a wonderful toy: but it’s not very much more.

Now, it certainly is fascinating. I know no one who has a Mac who wants to sell it. When it was my turn with the Mac that Dr. Hyson and I own between us, he parted with it reluctantly, and I can certainly see why. The machine is fun.

It just isn’t very useful because there’s no application software. Indeed, there’s less than we thought, because Microsoft’s Multiplan, which we bought with the Mac, is no longer being delivered; the dealers have just been told to take it off the shelf. We have not yet been told why or what Microsoft will do about our copy (which we paid full retail for. as we did with the Mac itself).

The Creator In fact, the only application software for Mac that’s actually on the market is Bruce Tonkin’s The Creator database. The Creator is a kind of personal filing system, and there are versions for nearly every microcomputer I know of. The data files created by it can be transferred among all these systems-except. of course, for the Macintosh, which sits in lonely isolation.

For the price—$35 postpaid, 30-day money-back guarantee—The Creator is one of the best values in micro land. I recommend it for nearly any machine; but especially for the Macintosh, since it lets you do something useful with the machine.

The Macintosh version of The Creator is written in Microsoft BASIC. It doesn’t use the Macmouse, because MacMicrosoftBASIC is too full of bugs to allow that. Instead, Bruce takes all input from the keyboard and formats it himself; of necessity, he has to treat the Mac as a glass Teletype, ignoring all its splendid Macfeatures such as Quickdraw, because there’s no way to get at them.

Tonkin writes many of his programs in his PBASIC, which is a preprocessor for Microsoft BASIC that works somewhat as the RATFOR precompiler works for standard FORTRAN. PBASIC lets you do structured code and handles most of the housekeeping for you. I’ve reviewed it before, and I still recommend it; Bruce has written some impressive software in PBASIC. The output of the PBASIC “compiler” is legal Microsoft BASIC; you can then truly compile that with Microsoft’s BASCOM to get tight, fast code that’s very portable.

Alas, though, it won’t work on the Macintosh, because MacWrite won’t let you build large enough source-code files to make it worth porting PBASIC over to the Mac.

For the Rest of Us?

I’ve lived in Hollywood for some years, and I’ve grown familiar with the typical Tinseltown deal. A producer goes to a star and says: “I’ve got a great script and your favorite director. Sure would like to have you in the picture.” Then he goes to the director. “I’ve got a star and a script, and the star sure wants you— Then to the writer; and finally to the money people. When it’s all finished, everything he’s said is retroactively true.

Similarly: Macintosh is going to sell like hotcakes because of all that wonderful application software. Now you can’t do much with software on the Mac, because there’s too little memory, and it takes anywhere from 5 to as many as 40 disk swaps to copy a Macdisk; but it’s all right, because there will be a second disk drive, and a hard disk, and other excellent hardware after-market add-ons to fit on the Mac’s “virtual slot”: and after that the software will be easy to write. Software houses are going to work very hard to write application software for the Mac because there’s such a huge market, since Apple’s selling an awful lot of Macs. Hardware houses will do Macadd-ons because there’s so much software. Etc.

It can certainly be made into retroactive truth. By the time this comes out, I expect it will be. Some of those Hollywood deals fall flat, though: it all depends on how quickly they can be put together before everyone in the industry catches on. Meanwhile, right now the Mac is mostly useful for people with special requirements, such as advertising layouts with text or producing memos with graphics.

It sure is fun, though.

Cataloging For some years now I’ve used Ward Christensen’s public-domain disk-catalog program. Lately, though, I’ve been importuned to try a new one called Eureka! from Mendocino Software. Eureka! has a number of advantages, including the capability to include lots of comments in your disk catalog.

so that you can figure out that “NASTYLET.TXT” was sent to your mother-in-law rather than your lawyer, or whatever. It also lets you date things.

Until recently, though, I couldn’t get Eureka! to work with my big CompuPro System 8/16, nor would it catalog the system’s hard disk. The Mendocino people kept trying, though, and eventually sent their stuff to Tony Pietsch; and as of an hour ago Noor Singh delivered a copy with the note that this time it works.

It’s about time I changed over, and the ability to add comments and dates makes Eureka! nearly irresistible. Full discussion next month, or Real Soon Now if too much flows into Chaos Manor in the next few weeks; meanwhile, Mendocino Software deserves some applause for plugging away until their people could make Eureka! work on my system. Thanks.

RAM Disk for the Z-100 Some weeks ago I got a letter from David lames at the University of Kansas. Zenith offered the faculty and students at UK one of those deals you can’t refuse, a Z-100 at a really good price, and Mr. lames bought one with a lot of memory. When he went looking for a RAM-disk program, though, he couldn’t find one.

I had my assistants send copies of his letter to half a dozen places known for their RAM disks. We got only two answers.

Zenith recommends Standard Data of Fort Lauderdale, Florida. I’ve not used this company’s boards, which come in 256K-byte sizes, but I suppose the hardware and software must work or Zenith’s people wouldn’t have sent me Standard’s address in reply to my inquiry.

I’d suppose that CompuPro’s M-Drive/ H boards could be made to work, since they’re IEEE-696 (S-100) standard; but I’ve never heard of an installation program that would get M-Drive/H running under Z-DOS. Someone would have to hack up a BIOS (basic input/output system) for that, and I don’t know of anyone who’s done it.

One other source I know of is Macro-tech out in the San Fernando Valley, which has a full 1-megabyte board for the Z-100. It’s been going for six months now with RAM-disk software.

Late addition: we have one. it works fine. More next month.

Eat Your Heart Out Macrotech had an impressive board running at COMDEX Winter: an 8-bit Z80 coupled with an Intel iAPX286 chip for 16-/32-bit processing.

It was installed in a CompuPro boat-anchor box much like mine. I admitted that I was impressed. To make sure there were no misunderstandings, Macrotech’s people opened the box to show me that the board was wire-wrapped: a production model will be available sometime this summer or fall.

The rest of the boards in the box were CompuPro. They claim that sometime before next fall’s COMDEX they’ll come over here, open my Dual Processor, remove the CompuPro 8085/8088 board, and insert a Macrotech board: after which my system will continue to operate on CP/M-8/16 as before, except that it will be faster, will be able to run Z80 as well as 8080 software, and will accept iAPX286 commands. We’ll see: my CP/M-8/16 isn’t the common variety, since Tony Pietsch did some work with the command processor, and enabled interrupts, and generally did a spit-and-polish job. On the other hand, I never throw anything away: I can certainly boot up the old standard CP/M-8/16 if 1 have to.

Printer Optimizer Last month 1 mentioned Applied Creative Technology’s-Printer Optimizer, which is a box full of memory that sits between the Golem (CompuPro 8/16) and the NEC 7710 Spinwriter. Shortly after that, ACT recalled the machine in order to add even more memory to it.

And I discovered that I was hooked. No sooner was the little jewel gone than 1 missed it terribly. It’s amazing how you can get used to the idea that printing is something that happens nearly instantaneously (well, at 9600 bps, which is the speed at which the computer ships data into the Optimizer). Waiting for the machine to print at normal speed (1200 bps) was just no fun at all.

Ten days later it came back. The first time it showed up I’d had Alex install it. but he wouldn’t be back from UCSD for a couple of days, and I wanted the Optimizer now: so I tackled the installation myself.

I am no hardware genius. Like most Americans of my generation, I learned to solder and install electrical switches and such, but when I was in school electronics meant vacuum tubes and hookup wire: not only didn’t we have transistors. but printed-circuit boards were pretty rare. Thus, while I’m not afraid of the innards of my computers, I do tend to think of hardware jobs with a distinct sense of unease, not to say dismay. There was nothing for it, though: if I wanted my Optimizer—and I sure did—I was going to have to do it myself.

It turned out to be simple. Not as simple as it would have been had I wanted to connect it to a Centronics parallel port instead of through an RS-232C serial port, but simple enough. The ACT instructions are quite clear, with plenty of diagrams and examples and a good explanation of the theory of what’s going on. It took me considerably less than an hour to hook things up, and mirabile dictu, everything worked first time.

Now, with 256K bytes of memory in the Optimizer, I can pack a great part of an entire novel into it. 1 can do that as one long file, or as a series of linked files, or even one file at a time if I want to fiddle with the last part while the beginning is printing out.

The Optimizer even has a way to program it so that the files aren’t necessarily printed in the order you put them in: there’s a way to shunt stuff off, as on a railway siding, so that something else can be printed first. It’s not a feature I use very often, but it can be convenient, as for example when you don’t have the fanfold tractor on because you want to print single sheets, but there’s a convenient opportunity to load in a file that will need fanfold.

If you don’t have a printer buffer, you don’t know what you’re missing. They come in a variety of styles with various features, and I haven’t much experience with any of them except the ACT Optimizer. I sure love this one, and it’s hard to imagine one easier to use. Given the wide variety of protocols and stop bits and other such stuff (the RS-232C meaning Revised Standard-232C, is anything but standard: there are a bewildering complexity of ways to hook serial ports to computers), it would be difficult to write better hookup instructions.

Recommended.

Speeding Up Your PC Another gadget I’d just got installed last month was the Quickon from Security Micro Systems. This little gizmo installs rather simply in your IBM PC: when it’s all aboard, all you’ll see is a small switch on the back.

Throw the switch one way, and the PC behaves normally. Throw it the other way, and the memory tests are disabled: the PC comes on nearly instantly. Now, I don’t think you ought permanently to disable the memory tests: but I sure don’t much care to wait for all that every time I have to turn the PC off to escape from some hang-up—and with no true hardware reset on the machine, that happens more often than I like.

We’ve had the Quickon working for five weeks now, with zero trouble: and Jim Baen reports that he’s had his almost a year. If you like waiting for the PC, you won’t need this, but if you’re as impatient as I am you’ll find it nearly indispensable.

Disk Maker New Generation Systems is a public benefactor.

Our Kaypro 4 will, courtesy of a program called Uniform—itself a lifesaver— read a number of 514-inch disk formats, but there are machines it has never heard of. However, when we get a disk that the Kaypro can’t read, we no longer despair: we take it to the Disk Maker I.

I first heard about the Disk Maker at CP/M East when I stopped to talk with Leor Zolman. Leor was, naturally, demonstrating BDS C, the blindingly fast 8-bit C compiler he wrote while he was still an undergraduate at MIT but he’d been so impressed by the Disk Maker that, although he owned no part of the company that makes it, he had one to show off at his BDS booth.

That was a pretty good recommendation by itself. 1 have considerable confidence in Leor’s judgment. He introduced me to the box’s inventor, and we arranged to get one shipped to Chaos Manor. The result has been as advertised. Disk Maker can read almost any conceivable 5-1/4-inch disk format.

The machine consists of one or two disk drives—mine has two, one 48 tracks per inch (tpi) like the IBM PC, the other 96 tpi like the Eagle 1600 series—and an S-100-bus disk-controller card. You can install the Disk Maker card in your S-100 system and forget it’s there until you need it; then, when you get a disk with a strange format, fire up the Disk Maker. Chances are very good that it will be able to read it, and you can use PIP to move files to and from your normal system’s disks.

We have the Disk Maker installed in Zorro, the Zenith Z-100. who also runs 8-inch disks in addition to his normal 5-1/4-inch disks; thus, we can move files from any 5-1/4-inch format to 8-inch IBM standard, after which they can be brought in for Zeke and the Golem.

The Disk Maker will format disks, too; about 50 formats are supported. It’s thus nearly ideal for a small software house that tries to support a wide spectrum of customers.

It’s easy to install, easy to use, and darned near unique. If you need one, you need it bad.

Raving About Calendar/i Peter Flynn, our new assistant, sees a lot of software; alas, more than I do, since so much flows through Chaos Manor that I have to let Peter and Alex screen it unless it’s something I’m particularly interested in. Thus, he doesn’t give many rave reviews.

Here’s one item he liked a lot.

“Calendar/1 puts date-related information into a calendar format. It is easy to use. The manual is good and organized in a straightforward manner. The preface states that the manual assumes you know how to use the various CP/M functions such as PIP and COPY and that you have and can use a text editor that can produce plain text files ‘with no embedded control characters.’ “The text editor is used to create one or more files that are the sources for the date-related information. This is the best approach 1 have ever seen for a calendar program; it is easy to input large numbers of notices, and it’s very easy to update and modify into different calendars. This is by far the most useful and versatile program of its type I have ever seen. It is much better than the scheduler in Valdocs.

“Calendars are composed of notices that are stored in calendar files. Each calendar file is composed of dates followed by notices pertaining to that date. The calendar files, which you create on a word processor, look like this: %This is a comment.

@ 8/10/84 Meeting with Fred at 7:15 AM 7/29 First Moon Landing. 1969 8/10 Sally’s Birthday “The order of the dates does not matter, and you can enter the same date more than once. They can be organized under headings such as birthdays, meetings, notices, and social events. Under each date is a notice or group of notices. They will be printed in the box for the corresponding date on the calendar. Lines beginning with % are comments and are ignored by the Calendar/ 1 program.

“A new calendar file can be for a month, a year, or many years. If a date ends in a year, such as 5/16/84, then it will be printed only in that year; if only a month and day are given, it prints the notice for every year. (Years can range between 1583 and 9999; if only two digits are given, the twentieth century is presumed.) “Calendar files can be separated or combined: you can have a file of birthdays and another of meetings, print a calendar for each, and combine those with others to make one master calendar. Calendar/1 comes with a number of prewritten calendar files, including holidays and historical dates of interest.

“Calendars can be printed on screen or the printer. You specify the length and width. If a notice won’t fit, what will fit is printed, an asterisk is added, and the balance is printed as a note on an overflow page. The layout’s good considering the space you have to work with.

“There are a number of other control features, all well documented.

“Calendar/1 is useful for scheduling work; many people can be given identical copies for job control; it is also useful for travel scheduling.”

We expect to make a lot of use of Calendar/1 here at Chaos Manor. It’s a well-conceived and useful program.

Out of Space Again …

There’s a ton of stuff on my list, and I’m out of space. At least let me mention the Infocom games, such as Sorcerer and Enchanter and the like. Not only do we at Chaos Manor love them, but I notice that my partner Larry Niven is hooked. For those few who don’t know, Infocom games are script driven: there’s no fancy graphics, no arcade action; only text adventures.

This kind of game grew out of the original Crowther and Woods Adventure of the Colossal Cave. The original Infocom implementers worked on Zork while at MIT they later developed other software for role-playing games. Some of the work they’ve done parsing and interpreting English is remarkable.

Enough. My taxes are due, after which I’m on the road for two weeks; meanwhile, I just opened a letter from Judy-Lynne Del Rey, my long-suffering editor at Ballantine Books; she’s expecting Larry and me to turn in Footfall Right Away: the letter said only, “Nag! Nag! Nag!”

I think that was a hint.

————————-
The Unshippable Sage

In my judgment, Sage computers have about the same place in the world of the Motorola 68000 and follow-ons that CompuPro’s boat anchor has in the Intel 8086 and beyond world: they’re very probably the best systems you can get for software development; they certainly have a lot of bang for the buck.

The Sage is also useful as a general-purpose computer: there’s getting to be a lot of good software, including engineering applications stuff, for 68000 systems. Sage uses a number of Sage computers internally for everything from accounting to new systems design; and therein hangs a tale.

When I visited Sage at the Reno headquarters I was shown through the whole plant, including the inner offices where new research is done. Sage’s people use quite a lot of modern equipment, including some fascinating stuff ‘for computer-assisted drafting and for chip
design. I also saw the oddest computer I’ve ever seen in my life. I mean. I’ve seen plenty of desktop computers, but I never saw one bolted to the side of a desk before; yet there was a perfectly good Sage II in that situation (see photo 2). It was running, too. None of the Sage crew seemed to think that was odd at all. They were used to it. Finally, I had to ask.

It turns out that in the early days of Sage, when they first began to ship machines, they got more orders than they could fill; so that whenever Bob Needham, one of the cofounders (with Rod Coleman), would get a machine to help him with advanced system design, someone would see it and ship it off to a paying customer. Eventually Bob decided that enough was enough and bolted a new Sage II, sans case and fan. onto the side of his desk. The disk drives and power supply were in a drawer.

No one has shipped that one…

fosdic III and the ’60 Census

Electronics lends a much needed helping hand to the census takers for this big once-in-10-years event.

IF you haven’t already, you’ll soon be filling in the decennial census form, 1960 version, designed to include every U. S. family. Nearly 160,000 persons will be employed by the Bureau of the Census to collect and process this data. The National Bureau of Standards has made their job easier by developing FOSDIC III (Film Optical Sensing Device for Input to Computers), an electronic workhorse that rapidly reads (detects) pen or pencilled marks in multiple choice answer areas that have been microfilmed. The information conveyed by these marks, scanned by an electron beam and photocell, is automatically coded into computer characters and then recorded onto magnetic tape. Thus FOSDIC III takes human information, identifies it, and converts it into computer language with electronic speed.

By The Year 2000,The World May Catch Up With The Way CompuServe’s New Electronic Mall Lets You Shop today.

Introducing the first computer shopping service that brings you convenience, savings and enjoyment Here’s your chance to expand the practical uses of your personal computer.

Sign up for CompuServe and shop in our new Electronic Mall. It’s easy to use It tells you more about the products you’re buying. It lets you order faster. And it’s totally unique CompuServe’s new Electronic Mall ” offers you all these shopping Innovations.

- It’s enormous! So it gives you in-depth information on thousands of goods and services, and lets you buy even hard-to-find merchandise. – Its unique “Feedback” service lets you ask the merchants themselves i specific questions. – It’s incredibly efficient in ordering the products and services you want – Its special discount opportunities make it economical, purchase after purchase. – And its name-brand merchants assure you of top-quality merchandise Make the CompuServe Electronic Mall 15-Minute Comparison Test What you can do in 15 minutes shopping the Electronic Mall way.

• Call up on your computer screen full descriptions of the latest in computer printers, for instance.
• Pick one and enter the order command.
• Check complete descriptions of places to stay on your next vacation.
• Pick several and request travel brochures
• Access a department store catalog and pick out a wine rack, tools, toys… any thing!
• Place your order.

What you can do in 15 minutes shopping the old way.
• Round up the family and get in the car.

The Electronic Mall, a valuable addition to the vast world of CompuServe.

CompuServe’s Consumer Information Service brings you shopping information, entertainment, personal communications and more.

You can access CompuServe with almost any computer and modem, terminal or communicating word processor.

To receive your illustrated guide to CompuServe and learn how to subscribe, call or contact…

CompuServe Consumer Information Service
P.O. Box 20212
5000 Arlington Centre Blvd.
Columbus, OH 43220
800-848-8199
In Ohio call 614-457-0802

MORE COLOR. MORE SOUND. MORE GRAPHICS CAPABILITIES.

Compare the built-in features of leading microcomputers with the Atari personal computers. And go ahead, compare apples and oranges. Their most expensive against our least expensive: the ATARI 400 Start with graphics capabilities. The ATARI 400 offers 128 color variations. 16 colors in 8 luminance levels. Plus 29 keystroke graphics symbols and 8 graphics modes. All controlled from a full 57 key ASCII keyboard. With upper and lower case. And the system is FCC approved with a built-in RF modulator That’s just for openers.

Now, compare sound capabilities. Four separate sound channels and a built-in speaker. With the optional audio/digital recorder, you can add Atari’s unique Talk & Teach Educational System cassettes.

Here’s the clincher: Solid state (ROM) software. For home management, business and entertainment. Or just plug in an Atari 10K BASIC or Assembler language cartridge and the full power of the computer is in your hands.

Memory? 8K expandable to 16K. And that’s just for the ATARI 400 at a suggested retail of only $549.99.

The ATARI 800 gives you all that and much more.

User-installable memory to 48K. A full-stroke keyboard.

With a high-speed serial I/O port that allows you to add a whole family of smart peripherals. Including up to four individually accessible disk drives. And a high speed dot-matrix impact printer. And, the Atari Program Recorde is included with the 800 system. Suggested retail price for the ATARI 800 (including recorder) is $999.99.

Make your own comparison wherever personal computers are sold. Or, send for a free chart that compares the built-in features of the ATARI 400 and 800 to other leading personal computers.

ATARI PERSONAL COMPUTER SYSTEMS
1265 Borregas Ave. Dept. C, Sunnyvale, California 94086. Call toll-free 800-538-8547 (in Calif. 800-672-1404) for the name of your nearest Atari retailer.

PC Owners… Reach for Your Phone! This Winchester is Loaded… with UNIX Software

That’s right partner. Now is the time to upgrade your PC with the Sundowndisk. Includes controller. Installs right inside your PC in less than 10 minutes. Backed by our full one-year warrranty.

But that’s only half the story . . . The Sundown comes loaded with VenturCom Venix/86. This highly-acclaimed operating system is a licensed implementation of AT&Ts UNIX and is the only MULTI-USER, MULTI-TASKING UNIX environment available on the IBM PC. Plus you can store and run your MS/DOS programs and files as well!

We offer immediate delivery. And our price . . . now that will blow your boots off! Need we say more? Reach for your phone and dial: 617-491-1264
Unisource Software Corp., Department 4109
71 Bent Street Cambridge, MA 02141

view additional pages

Undo. Windows. Mouse. Finally.

New Microsoft Word. It makes your IBM Personal Computer think its better than a $10,000 word processor With Microsoft Word, what you see on the screen is what you get on the paper So its easy to spot mistakes. Boldface, underline, and italics look like this, not this: ^Bboldface^B, ^Sunderline^S, ^Italics^I And, when you make changes, paragraphs are automatically reformatted. Flush right, flush left, centered or justified. It even gives you several columns on a page, like a newspaper.

Word forgives and doesn’t forget.

There’s an “uh-oh” command called Undo. Make a mistake? Or just want to experiment? Hit Undo.

Word undoes your last change and remembers things the way they were.

Word does windows. Up to eight, to be exact. So you can transfer or edit between eight different documents. Or between eight different pieces of the same document

Word travels fast.

Word has a Mouse, a handy little critter that lets you move copy, select commands and edit faster than you can say “cheese.”

Word also lets you create your own style sheets, so “you can standardize your documents, memos, files and letters.

It’s not surprising that Microsoft has a way with Word. We designed the MS™ DOS operating system that tells the IBM® PC how to think. And we pioneered the first microcomputer BASIC, the language spoken by nine out of ten micros worldwide.

For a few final words, call 1-800426-9400 (in Washington State call 206-828-8088) for a free Word brochure and the name of your nearest Microsoft dealer.

MICROSOFT
The High Performance Software

view additional pages

THE WEST COAST FAIRE

by Jerry Pournelle

The lord of the Manor visits his favorite computer show.

The West Coast Computer Faire is my favorite computer show; how can a big publishing company like Prentice-Hall put it on the way Jim Warren did?

Of course it can’t and it didn’t, but in justice. Jim Warren wouldn’t have been able to keep it up. either. For better or worse, the micro industry has changed. Oh, sure, there are still some pretty good products for sale in the little booths along the walls, but there were fewer than in the old days for the simple reason that the big outfits also have a lot of incredible new stuff, and it’s a lot harder for a newcomer to compete.

This Faire was to be four days long, with Wednesday as set-up day. I went up Tuesday evening to spend some time with some of the BYTE editors. Wednesday morning I went down to the Faire office. The press room wasn’t set up, but I hadn’t expected it to be. I went to exhibitor registration and asked for a press badge.

They’d never heard of press badges. Hah, thought I. Changes. Warren always let us get in early. Set-up day is a good time for interviews and private demonstrations before the crowds arrive. These people don’t understand that. Ruining the Faire. they are…

Someone sent for David Sudkin. He listened for a mo- ment, said “Sure!” and produced a big red “TEMPORARY” badge. End of problem. I wandered in to talk to people.

As usual, set-up day is a study in controlled chaos. The big companies like MicroPro, Apple, and IBM have professional set-up crews working with the local San Francisco exposition services people. The smaller outfits are on their own.

At least they used to be. This year the unionized exposition troops were greedier than usual In the past they’d let a small booth owner carry his own stuff; this year they socked everyone for a couple of hundred bucks to let them unload their trucks. It’s a good racket: $200 to carry 10 boxes a hundred feet or so.

Eating the Seed Corn.

It’s also a stupid racket, like eating the seed corn, because some of the smaller outfits won’t come back. Indeed, all day Wednesday the talk among exhibitors was that it was time to get the Faire out of San Francisco and into a friendlier city. More than half the exhibitors I talked with swore they wouldn’t be back next year.

There were a lot fewer, anyway. That is; there was a lot of space rented. Not quite as much as last year: no one was on the garbage-collection platform, and the chair-storage area, otherwise known as the Black Hole of Calcutta, hadn’t been rented, but everything else had been, including the two side halls that used to hold seminars and speeches.

Instead of lots of small exhibits, there were fewer, but much larger, setups. Three of the biggest were in a row: IBM, Apple, and Digital Equipment Corporation (DEC).

Wednesday ended with the traditional “after set-up” party thrown by Compute! magazine. As usual, there were more press people than exhibitors snarfing up the hors d’oeuvres and free beer. We also noted that there was no exhibitors’ lounge: that location had been rented by Apple to let people get hands-on experience with Macintosh, or maybe it was IBM doing the same thing with PC and PCjr. The party was in the concession area—and unlike last year, there were no chairs and tables, just food service counters, with booths across a narrow aisle from them. I pitied the people who had those booths.

When I left on Wednesday night, it was a scene of pure chaos. On Thursday morning, the miracle had happened again: everything was set up. I headed upstairs to the Workman and Associates booth for coffee.

No coffee. The union people had not only forced Barry to pay for work he hadn’t needed, but they’d lost the box containing vital parts to the coffee maker. Mornings are not endurable without coffee: you might as well sack a city. Eventually I found some in the press room, but by then I was thoroughly prepared to make this my last Faire.

Enter the Populace.

At 11:00 the doors opened and the crowds came in.

There were quite a few people. Not as many as we had on the first day last year, but a respectable number— and this was, after all, Thursday. Last year the Faire was only three days and opened on Friday.

They were interested people, too. Fewer hackers—but a lot of customers. As I made my hurried cruise through the aisles, it seemed that more people were buying things. Barry Workman confirmed that when I got upstairs. There were a lot more serious customers.

All the presentations were well attended, too. The panels for hackers and hobbyists had plenty of those, but the panels for newcomers and users had crowds, too. The “All Star” speeches, which featured Bill Gates, Gary Kildall, Bill Godbout, Steve Wozniak, Adam Osborne, and, somehow inserted into this rarified atmosphere, me, were all crowded with attentive people who asked intelligent questions. This was a serious crowd.

By Thursday night it was pretty well confirmed: there were some glitches and justified complaints, but the new staff was trying hard to take care of the problems. This was a different Faire, but it was still a good one.

Barry Workman sells a line of CP/M software, including WRITE, my favorite text editor. He had more sales the first two days than he did during the entire Faire last year. Others report the same, and that’s not just software vendors. The people at the Priority One booth had lots of business, and when they announced a big sale on CompuPro 40-megabyte hard disks (like mine), there was something approaching a feeding frenzy. Even vendors of computer furniture reported big sales.

I expect there are several reasons for this. The state of the economy didn’t hurt, of course. More important, though, more people have microcomputers, so they’re ready to buy software and accessories. Whatever the reason, they were buying, and by Saturday night I could tell the exhibitors by their satisfied smiles.

Some Hits.

There wasn’t much new and exciting stuff. The best hardware wasn’t at the Faire at all: Richard Ohran, of Modula Computers Inc., was on a panel about Modula-2 with me. He brought a couple of the U.S.-built modifications of Niklaus Wirth’s Lilith computer, but they could only be seen in his hotel room. Naturally, I went. It’s wonderful. It has fabulous graphics capability, better than Macintosh or even the (alas, late and lamented) Diser Modula Engine, except that the Modula Computer Lilith has a Winchester hard disk and a 5 !4-inch floppy disk—much nicer than the goofy Honeywell-Bull “washing machine” hard-disk cartridge that came with the Diser.

The Modula-2 panel was interesting, if a bit frustrating. Panelists were Ohran, Tony Gorrengourt of Logitech, John M. Craig, formerly of Diser and now with Ohran’s company, and Joel McCormick of Volition Systems. I’d been asked to be chairman: since I was going to the panel anyway, this gave me a good seat, although I didn’t have much to say. We had a good crowd, every seat taken: there’s a lot of interest in Modula-2.

Tony Gorrengourt led off with a neat presentation on the philosophy of Modula-2. Then John Craig described the U.S. version of the Lilith machine. Joel McCormick promised that Volition Systems would get the 68000 native code Modula-2 compiler finished Real Soon Now. Richard Ohran described the public-domain Modula Research Institute compiler (for the IBM PC), which is available for $40. (Useful if you’re interested in Modula-2: alas, not really in shape for doing much practical programming.) We were also able to announce that in early March a group met in Zurich to form the International Modula Users’ Association, and we even agreed on some standards for the language and library modules. With any luck, we won’t have a proliferation of dialects and libraries the way things went for Pascal.

Turbo!.

One very hot item was Borland International’s Turbo Pascal, which was on a show special at $46.95, not much under the standard $49.95, but low enough to attract crowds. Borland’s president, Phillippe Kahn. was there to demonstrate his product and to talk with anyone who’d listen. He’s planning a whole line of products, all to be sold at under $70. Unlike most software publishers, Kahn tries to see how much he can include in his packages. For example, the PC version of Turbo Pascal comes with a nifty spreadsheet program—complete with source code. The source contains a number of screen-handling procedures I’d have been willing to pay separately for.

I’m pleased to say Turbo was selling like mad. In case anyone has any lingering doubts, let me repeat: Turbo Pascal is as good a Pascal compiler as can be found on the micro market regardless of price. The new Turbo 2.0 for the PC has overlays, allowing you to write big programs. The integrated editor is a feature found only in really expensive compiler systems. Turbo Pascal is, as I write this, the only Pascal compiler that will work with the IBM PCjr. Not even IBM Pascal will do that! Besides, Turbo is much better than the Pascal IBM sells.

Incidentally, Phillippe Kahn tells me that he tried to get IBM to market Turbo, but IBM wouldn’t do it unless he raised the price to at least $200. For reasons he doesn’t understand, he hasn’t been able to find anyone at Apple who’ll talk to him at all about getting it out on the Macintosh. (Borland doesn’t yet have a Macintosh Turbo. Pity.) Turbo Pascal is an example of where I see this industry heading: an excellent product, well documented and well supported, sold at a fair price.

Blaise.

One reason many of us like Pascal as a programming language is that you can build up a whole raft of utilities—a box of software tools, if you like—and make use of them in many different programs.

Blaise Computing—I suppose the name is derived from Blaise Pascal?— offers a set of functions and procedures for writing programs in either Microsoft or IBM Pascal. They include both string handling and screen handling. There’s a distinct possibility that using these could save a lot of time: it would depend on just how well done the tools are.

They don’t support Turbo, alas, but source code comes with the tools, so it shouldn’t be that hard to adapt. I haven’t actually looked at what they offer, so I can’t tell how much they go beyond the routines given in Peter Norton’s book.

Friends and Books.

One reason I like the West Coast Faire is that I meet a lot of old friends and make new ones. This year I met Doug Clapp, who has been, uh, kind enough to mention me in his columns. Clapp’s book Macintosh] Complete was, I think, the first one on the Macintosh to be published: alas, it reads as if it were rushed into print, and most of what’s in it is either elementary or full of goshwow! breathless awe of the machine. It wouldn’t be a bad beginning book, except for the price, which is pretty steep for what’s actually covered.

A better Macintosh book is Cary Lu’s The Apple Macintosh Book, which goes into heavy technical detail but also covers a lot of elementary essentials. Like Clapp’s book, it assumes there’s going to be a lot of software for the Mac and that many of its problems will be solved. If you have to get your book out the same month that the computer appears, you’ll always have this problem, of course. Eventually there will be books written by people who have used the mature system: until then, Lu’s is certainly the best reference I’ve seen, and I recommend it to anyone who either has a Mac or is thinking of getting one. Alas, Clapp has a really bad case of Macworship: you won’t get an objective assessment from his book.

Robert Tinney, the genius who does the BYTE covers, was at the BYTE booth with Susan, his wife. He was selling posters made from his old covers, and doing a pretty good business. He also took some photographs of me: he’ll be doing the cover to my collection of columns and reminiscences that comes out this October from Jim Baen Associates.

While I was talking with Robert and Susan, I met Peter Norton, whose Inside the IBM PC is just plain excellent. The book gives lots of examples, in BASIC, Pascal, and even assembly language, but it’s written in a clear and intelligent style that neither talks down to the reader nor leaves out important information. Highly recommended for anyone with a PC or close PC clone.

Infoworld. as usual, held a big blowout party: this year it was in City Hall. I ran into John Dvorak there and couldn’t resist asking him to rate the party, but he wouldn’t do it. Adam Osborne was also there. He’d just announced his new software development and publishing venture. He seemed happy enough. Alas. I couldn’t stay long: the party featured a rock band and was held under the City Hall dome. Great acoustics if you want to be deafened…

Dimension.

One of the most interesting hardware items was Micro Craft’s Dimension 68000 “Professional Personal Computer.” At CP/M East in Boston these folks had only one copy of the machine in their booth: now there were half a dozen, and they were selling them right there at the show: I noticed a couple of software companies buying, and Chris Rutkowski’s Rising Star, which is overhauling the Epson QX-10 software, seemed decidedly interested.

The Dimension is supposed to run software written for nearly any 8-bit or 16-bit machine, including Apple, IBM PC, Kaypro, TRS-80, UNIX, and Osborne. The idea is that Dimension owners won’t have to worry about software compatibility: one machine runs everything.

I’ve had no chance to examine the machine, so I merely report the company’s claims. Micro Craft had one set up with transparent cover, and it looked well constructed, but that kind of inspection isn’t worth much. It’s certainly an intriguing concept.

The Big Mac.

The most spectacular display at the Faire was Apple’s giant-size model of a Macintosh. This big Mac is large enough to walk into! The screen is a projection image of a nearby normal size Macintosh, so that crowds of people can watch Macdemonstrations, and it certainly drew the crowd.

There were also several dozen Macintosh computers upstairs; those interested could make an appointment to go play with one for half an hour. Many people took Apple up on the offer, and they all seemed to come away with smiles. There’s no doubting it: the Macintosh is fun. I haven’t seen so much excitement since the very earliest days.

In spite of all the Macexcitement, there was essentially zero applications software for the Mac. No one had any hardware to hang on the Mac’s I/O port (called a “virtual slot” in Macsales literature). There was a MacForth to be available Real Soon Now, and the FORTH people had a second disk drive—without a case—attached to the Mac. They’d apparently cobbled that up themselves; Apple wasn’t supplying second drives. Or perhaps they were blessed.

Macintosh comes with Macwrite. a limited text editor, and Macpaint graphics-support packages. Apple expects outsiders to develop software for the machine. So far there isn’t much. Except for Microsoft’s Multiplan—which is copy protected on the Macintosh although the IBM PC version is not; presumably Macintosh users are less honest than IBM customers?—I saw only one other applications program for the Mac.

That was at the Workman and Associates booth: Bruce Tonkin, software author, shared Barry’s booth. Bruce writes really complex programs in Microsoft BASIC (MBASIC). Actually, he writes in P-BASIC, which is a structured BASIC precompiler that compiles to, of all things, legal MBASIC. The P-BASIC compiler is written in Microsoft BASIC, so it wasn’t very difficult for Bruce to get it onto the Mac.

He used it to write a highly complex database program called the Creator for the Macintosh. Guy Kawasaki, Apple’s Macproject head, was quite impressed when he saw it running. It’s a good program, not up to dBASE II, but good enough for a lot of serious business work, and it’s fast (even in interpretive Microsoft BASIC).

That was all we found for the Mac. The machine has some limits, especially for serious business users. Lack of applications software is one major difficulty. A full discussion of this machine and what it means to the micro world will take more space than I have here and appears in this month’s column.

More Networks.

Bill Godbout’s CompuPro company was proudly showing off the four-user Shirley; you’ll recall CompuPro had a mock-up of it at the last Faire. Shirley puts four computers in a single box and internally networks them; each user has a central processor and memory banks.

The CompuPro folks announced that they’ll be using Datapoint’s Arcnet system to tie their multiuser machines together. This includes not only the Shirley machine (officially known as the “10″) but also the new machine based on the iAPX20286 chip. The 286s are pretty expensive just now, but they are available to software developers.

CompuPro also had Concurrent CP/M 3.0 running on an 8/16 8085/8088 Dual Processor system much like mine. With four users and eight tasks, it worked, all right, but it was pretty slow. I’m much looking forward to getting Concurrent CP/M with windows running here, but I’ll never put multiusers on a single microprocessor. One user, at least one central processor; that’s the rule (mine, anyway).

Over the Rainbow.

Digital Equipment Corporation had a large booth just next to Apple, but the people there seemed to have trouble generating any excitement.

DEC is a company that has learned fast. When the people at DEC first brought out their Rainbow PC, they seemed determined to isolate their users from the rest of the micro world. The Rainbow wouldn’t even format its own disks! You had to buy them Preformatted.

No more. Rainbow now reads lots of disk formats and runs all manner of outside software: DEC never tires of bragging about just how much. Like the Fujitsu, the DEC Rainbow has both 8-bit Z80 and 16-bit 8088 microprocessors and can run CP/M-80. CP/M-86, and MS-DOS software. It’s a pretty machine, with a good display—and very nearly the same wretched keyboard as the IBM PC. Since the IBM designers have said in interviews that they just sort of “made up” the IBM PC keyboard, I wonder why DEC’S salespeople insist theirs is some kind of European standard?

The Rainbow deserves more attention than it got: being next to the overcrowded Big Mac display must have been pretty grim for the DEC booth people.

Corvus.

As I’ve reported in previous columns, we have a Corvus Concept computer. It implements Corvus’s Omninet system, which lets us link up an Apple, IBM PC, Zenith Z-100, TI PC, and CompuPro S-100 bus system so that they can’all share the Concept’s large hard disk.

The only person I “know” at Corvus is a software engineer named David Ramsey, and I’ve never actually met him: we’ve corresponded and talked many times on the phone. When I spotted the Corvus booth at the Faire, I headed right for it. A rather pretty young lady with blue eyes was the only person visible.

“Is David Ramsey here?” I asked.

She didn’t hesitate a second. “Oh, no. We don’t let him come to the Faire. We keep him chained to his desk.”

Alex was later told the same thing. This makes me more anxious than ever to meet him.

Corvus has made some changes in its system. For example, our Concept’s hard-disk unit is very large, about 1-1/4 times as wide as an 8-inch disk drive and a little taller. This size is a hangover from the days when they shipped 8-inch hard-disk drives some years ago, since the drive inside is a 40-megabyte 5-1/4-inch drive. Corvus showed off a new hard-disk drive at the Faire. It’s only a little larger than the 5-1/4-inch hard disk it contains.

The new model has the Omninet interface built in—ours has a separate unit, about 1-1/2 times as high as a half-height 5-1/4W-inch disk drive, to connect it to the Omninet. Believe me, the amount of space taken by these drives is large, so I was happy to see these improvements. Mr. Ramsey swears he’s coming down to Chaos Manor to oversee installation of the new and improved system, run twisted pairs of Omninet wires around, and in general improve on communications here. I’m much impressed by the Corvus; I keep hoping the company will mate that big bit-mapped screen to the Modula operating system, but Corvus says it has something even better, or at least as good, which Ramsey will show us when he comes. I’m prepared to be impressed.

1 just hope they take the chains off before they let him come.

PC-Write.

PC-Write is a full-screen text editor that makes use of many special features of the IBM PC. I haven’t tested PC-Write, but I’ve seen enough of it to know it’s a pretty good editor—especially for the price. You can get it for $10, or even for free.

PC-Write is distributed as “Shareware.” What this means is that you can buy a copy from the publisher for $10, but that’s not the principal means of distribution. Anyone who has a copy is encouraged to give copies to all his friends. The program and its documents are all on disk, so this is quite feasible.

People who get the program and like it are requested to send $75 to Quick- soft, the program’s publisher. When you send in the money, you get a printed copy of the document—a marginal improvement over what’s on the disk—and an “official registration number.” Put the registration number into your copy. Now, when you give copies away, if anyone you’ve given it to sends in the $75, you’ll get a $25 sales commission.

Quicksoft’s Bob Wallace says the company’s not doing too badly. A number of people have liked the program enough to register it. He’s meanwhile writing new programs to distribute this way.

I don’t know whether the “shareware” concept will catch on. I encourage the trend.

Tutsim.

One program I’m likely to use often has the odd name of Tutsim and comes with a picture of an Egyptian pharaoh. I presume he’s King Hit. I have a copy, and when I get a chance I’ll wring it out; meanwhile, I saw it work at the Faire, and it was impressive.

Tutsim solves differential equations. It does this by turning your computer into a kind of analog device: that is, you can simulate the various kinds of operational amplifiers some of us older types remember from days long ago. Once you have the analog model set up, Tutsim solves the equations and produces the graphical results.

This probably makes little sense to some readers. Don’t worry about it. I’ll have a lot more in upcoming columns. Tutsim looks to be a very practical engineer’s toolbox, and anyone who has to work with, or teach, differential equations should write Applied i for more information. It might be exactly what you’re looking for: certainly it was one of the most unusual items I saw at the Faire.

Rana Systems.

I don’t recall Rana Systems being at the Faire before, but I may simply have missed it. Since the last show, I’ve met Fran Mulvania, Rana’s director of communications, and we have installed Rana systems, so I tend to notice the company more.

Anyway, the Rana folks were out in force this year with a variety of products. Their flagship item remains the 2.5-megabyte floppy disk that so im- pressed me last year at NCC. They also make a full line of disk drives for Atari and Apple computers. I’ve used them, and they’re faster and more reliable than the ones Apple and Atari supply.

Rana’s monster floppy-disk controller goes into a standard IBM PC slot. It can control standard floppy disks as well as 2.5-megabyte drives (good for an overcrowded PC). I’ve been using 1.1-megabyte 8-inch floppy disks for some time now, and every time I have to work with smaller disks I miss the big storage capacity. It would be a weird experience to have 5-1/4-inch disks with a greater capacity than my 8-inch disks!

Rana will shortly have the 2.5-megabyte disks for Apple IIe systems as well. We expect to get one pretty soon. Full report then.

Osborne.

Suddenly a lot of people were wearing buttons that said “Osborne is back!” However, if you read the fine print it wasn’t so certain.

The people at Osborne had a booth, not as large as the ones they had in the salad days, but large enough and complete with balloons. (Interestingly enough, Kaypro. which is the entry-level system I most often recommend, was in the booth Osborne occupied last year.) The Osborne people had videotapes of Adam Osborne himself telling how wonderful the Osborne computer is. They even had a bunch of computers that looked like the old Executive, until you got close to them.

What they didn’t have was order books. No computers were for sale, and no price for the new Osborne PC has been established. This Osborne is different enough from the ill-starred Executive to warrant calling it a new machine. They say it is definitely PC compatible: I saw it run Flight Simulator, which is the program people often use to test PC compatibility. Otherwise, the machine looked a lot like the older Executive.

There are a lot of differences, though. There’s an RGB (red-green-blue) color board adapter, an expansion chassis, and a socket for the 8087 math chip.

At last year’s Faire. Osborne had patches with the big “barred O” symbol and the words “Osborne—teaching others to be humble.” There weren’t any of those patches this year.

Zenith’s Friends.

As usual, Zenith took one of the largest booth areas of the Faire. It doesn’t look all that big, because the folks at Zenith always invite certain people to share it with them. For example, Jim Hudson, whose 8087 math board greatly increases 8086 machine capabilities, was there. Jim was showing off his 8087 boards for the Z-100 and the CompuPro. The Z-100 board is especially ingenious: it’s an S-100 board with 256K bytes of RAM (random-access read/write memory) and the 8087 both aboard. The 8088 goes on his board, and a header plugs into the motherboard where the 8088 went: the memory connects to the S-100 connector and gets all the decoding and so on there. The Hudson board has a lot of bang for the buck.

Walt Bilofsky’s Software Toolworks also shared quarters with Zenith. The Toolworks has changed a bit. Instead of Insta-press flyers, Walt has a big color catalog printed on slick paper. The logo of the blacksmith forging a floppy disk on an anvil is gone, too. For nostalgia buffs, he still distributes software in Ziploc sandwich bags though.

The Software Toolworks used to specialize in Heath/Zenith software, and the company still has a lot for those machines, but most of what I got from Walt this trip is for the IBM PC. I’ll report in a later column: meanwhile, those who don’t know about the Toolworks ought to get a catalog. Bilofsky sells good stuff at remarkably low prices.

The MPI printer people were also in the Zenith area. Zenith now sells MPI printers. So does Heath in its catalog. The printers say “Zenith” on the outside, but the catalog lists them as MPIs. Pretty honest. I’ve written about the line before: we have four MPI printers, and we’re happy with them all.

Heath/Zenith’s New Z-150.

The Zenith Z-150 ZPC is its entry into the IBM-compatible arena, and it’s an impressive one, so much so that my first thought is that this is probably the PC clone to buy if you want one. We’ll be getting one shortly—full report then. It sure has a lot of nice features.

The Z-150 is a desktop, smaller than the IBM. It seems to run everything that the IBM will, right out of the box. We watched Bruce Tonkin put his MyWord word-processing program in the Zenith and run every feature. Rachel Klau, a programmer for Heath, said that people came by through the entire show and ran all their favorite IBM programs without fuss, muss, or bother. While she was saying this, a group of people came over and ran their newfangled 2-D structural modeling program on it. They all applauded when it ran and went away saying that it worked better than on the IBM PC. After Rachel had relaxed (they were still batting 1.000 on IBM compatibility), she told me that the Z-150 was designed from the start to run Microsoft’s Flight Simulator, which is in some ways the very best test of IBM compatibility.

It’s certainly the only IBM compatible that you can build yourself. If you’re skillful with tools and soldering iron, you could save yourself a lot of money. The folks at Zenith showed a prototype of their “portable” (33 pounds!). Alex says it’s one of the funkiest machines ever built. It’s portable in about the same sense that a 19-inch TV with a handle is portable—with great care you can get it from room to room, lb be fair, the Compaq weighs 32 pounds and the IBM portable 30.

Both the table model and the “portable” have eight IBM-type slots and an 8087 socket, and both run at the same speed as the IBM PC. Given the company’s quest for absolute compatibility, that’s no surprise, but there is no speedup switch or anything to let the Z-150 cut loose on its own, and that’s a bit sad. Still, the machines seem to run absolutely everything that people tried, including MS-BASCOM (Bruce wrote MyWord in BASCOM). The Z-150 uses two half-height 5-1/4-inch drives.

You can get it with a hard disk, which takes the place of one of the floppy disks. Since it’s a full-height hard-disk drive, it sticks down into the cabinet below where the lower half-height drive would normally go. Very slick. The detached keyboard doesn’t have quite as nice a touch as the Z-100′s keyboard, but it’s better than most. The keyboard has moved all the “wrong” position keys on the IBM PC to more reasonable places. It has lighted LEDs (light-emitting diodes) for Caps-Lock, Num-Lock. and Shift-Lock keys.

This is definitely a “Heath/Zenith” machine, in that order. It was developed in St. Joseph, Michigan (Heath’s headquarters) by Heath employees. The Zenith Z-100 was developed by Zenith, though of course Heath is famous for its kit versions of computers—including the Z-150 ZPCs. Kit versions usually bear the Heath name and fully assembled ones the Zenith name.

The Z-150′s main claim to fame is its color board. It seems 100 percent compatible with IBM color programs and is by far the best PC clone color board I’ve ever seen. If the folks at Zenith want to sell this board to IBM owners, they’ll make a mint. Unlike the IBM, which blinks horribly every time a line is scrolled, Zenith’s has (settable!) smooth scrolling that looks good with both graphics and text. In addition, I heard rumors of high-resolution color boards in development. Don’t be surprised if the first one looks a great deal like the Z-100′s color palette.

One other twist: there’s a built-in debugging tracer in ROM that you can access at any time. It shows you what code is being run, what’s on the stack, and what’s in the registers. You can get to it as long as the machine is listening for interrupts; if not. you’ll have to turn it off and on. The lack of a real reset switch, one that always works, is the major blemish on an otherwise nearly perfect machine.

Incidentally, don’t expect Zenith to suddenly phase out the Z-100 in favor of the Z-150. This is the same Heath that will sell you a knob for a 2 5-year-old kit tuner out of its warehouse stock. If the company’s reputation won’t convince you. the sales of the Z-100 should: GSA, the Army, Navy, Air Force, and UPI have all bought more than 5000 each of these machines. It’s likely to be the machine at the Air Force Academy. The Z-150 is to get Zenith an entry into the market of PC clones, not because Zenith wants to dump the Z-100. A small company looking for a machine to develop PC software on could do a lot worse than buy the H-Z150 kit.

Speaking of the Navy, I talked to two sailors, one from the Enterprise, the other in an LHA helicopter carrier (or Large Hotel, Afloat). Both had Z-100s onboard—there are 23 on the Enterprise. Both complained that the power supplies tended to burn out on the Z-100s because of the spikes in shipboard power. Otherwise, they are extremely happy with their Z-100s. Considering that most big shipboard navigation and fire-control computers have just been upgraded to 64K-byte core (!) memory, I can see why! It isn’t generally known, but the BB New Jersey still uses big mechanical analog integrators for fire control. A Z-100 with a good power filter would be a lot more machine….

Bottom Lines.

There was a great deal more to see. including a Japanese import color system using a light pen to do in 265 shades of color what Macpaint does in black and white. The FORTH people handed me a thick stack of books and program disks that are supposed to produce the brand of FORTH described in Leo Brodie’s excellent introduction. Starting FORTH. MicroPro has a new 16-bit spelling-correction system called CorrectStar. It has more features than The Word Plus, which I use at present, and I liked it quite a lot; I’ll go into more detail on that in my next column.

There were a lot of good speeches and seminars, all well organized. Despite a few glitches, the new Faire management team kept things running far more smoothly than I would have believed they could. Final attendance was a bit over 40,000, down from the 48,000 of last year, but good enough, especially since so many people bought merchandise.

It’s still my favorite computer show.

IBM’s “Channel Arrangement”

I ‘m not much impressed with the IBM PCjr, so 1 would have ignored the IBM booth, but the folks there had big signs proclaiming their new “network.” This looked interesting, and 1 sent my son Alex off to investigate. Here’s his report: “IBM showed off a new product to tie the PC. PC XT. PCjr, and Portable PC together. It’s not a network. It’s a ‘channel arrangement: I’m not sure what that is. I can tell you what it is not: a real network. With a raw communication speed of 375K bits per second (bps) and only 64 users maximum, it’s not going to compete with Ethernet, Omninet, or Arcnet. lb compare. Arcnet and Omninet run at about 1 million bps and Ethernet at about 10 million bps. Those are raw speeds, not equal to the amount of data that can really be sent. Nonetheless. I think the low speed alone means it won’t do well, even with the IBM name on it.

“Data is carried on a 75-ohm coaxial (TV) cable. You put a Tee in for every station; that Tee cable can be about 16 feet 15 meters) long. The distance between the farthest stations can be only 1000 meters at most, so IBM’s channel arrangement’ is good only for pretty local communications to start with.

“There are some other problems. IBM doesn’t even support all of its own machines: 1 specifically asked, and no. the company doesn’t support the PC XT/3 70 on this net. Nope, no way to hook up to its own mainframes or non-PCs. Only 10 megabytes can be shared on line: the folks at IBM don’t even support multiple hard disks! They said you cannot put a second XT hard drive on the channel.

“There’s no true file sharing or record locking (goodbye, shared databases). No DMA (direct memory access), which means that each byte must go through a register (rather than being transferred directly into the receiving machine’s memory). This omission is ostensibly because the PCjr doesn’t support DMA. (I think that last point is a nail in the coffin of both the jr and the channel arrangement.) “The IBM employees I talked with stressed how you could run the PCjr with no disk drives by doing an ‘IPL’—an old IBM mainframe term meaning Initial Program Load—from the net. making the PCjr a diskless workstation. Perhaps they are attempting to crack the school market with this? This might work if it didn’t require a hard disk, but in fact the ‘channel arrangement’ does require one. I could see a bunch of PCjrs being fed by a single PC XT. but I see no cost savings over PCjrs with a disk. First, you must buy IBM’s board and software ($400 in quantity), then you must buy a central PC XT ($5000 and up). A PCjr’s extra memory and disk drive are only $600: that’s an awful lot of PCjrs to buy to get a cost savings—and there’s very little school software for the PCjr right now.

“But: IBM knows all these things. Rumors have it that this isn’t IBM’s network. but merely an aberration (er, stopgap) to fill in at the moment. All those deficiencies lead me to believe that the folks at IBM are using their customers as paying test sites to de-kink their ideas of what a network should be. ”

—Alex Pournelle

Hand-held Microwriter

If you can’t type, yet want to write perfect letters or memos without the help of a secretary, Microwriter could be the answer. It resembles a large pocket calculator, but has only five main keys, which fit the relaxed finger positions of your right hand. Individual alphabet letters are formed by an easily learned finger code, in which one or more keys are pressed for each character.

Your words come up on an LED display panel showing 12 characters at a time, and scrolling to the left as more are added. The full text is stored in a RAM (random-access memory) that holds the equivalent of eight normal pages of typescript. You later plug the

device into an electric typewriter that taps it all out at over 500 words per minute.

This battery-powered word processor is the invention of Cy Endfield, an American author and film director living in London. He says the simple chord keyboarding can be learned in half an hour, and that in a few days a novice will be “writing” faster than longhand. As memory aids, finger combinations are logically chosen to resemble alphanumeric shapes.

A sixth, thumb-operated control key pressed in combination with others is used for numbers, capital letters, punctuation marks, and other keyboard symbols. It also handles 16 editing functions, including backspace, insert, and delete. You can make changes and corrections as you go along, and end up with perfect finished copy.

Microwriter is “an electronic substitute for the fountain pen,” says End-field. He thinks engineers, office workers, and journalists will find it practical. In a large office, up to 15 units could be served by one automatic printer. Easy typing for the blind is another possibility.

At present, the system is distributed by rental solely in Britain.—D.S.

the case of the disappearing data

There’s Norman L. Battle, who needs the information on a computer tape in a hurry.

There’s Nancy Havens, who went to the tape vault and was never heard from again.

There’s Horace Cotton, who thinks he saw the tape in the hands of Don Walters, but won’t swear to it.

There’s JoAnn Taylor, who saw Don Walters whispering mysteriously to Marilyn Smith, by the water cooler.

There’s Marilyn Smith. (Can you blame him?) But where’s that tape?

It’s elementary. An Acme Visible systems analyst is what’s needed here. He may not find that tape, but he’ll darn well make sure it never happens again. He’ll analyze your information retrieval problems and come up with a system that makes all data instantly accessible, instantly visible. All data . . . computer print-out, filing, billing, work orders, inventory or whatever information your business depends on. If it’s worth keeping, there’s an Acme Visible system to keep it instantly accessible. And there’s a nearby Acme Visible systems analyst listed in your phone book. Now, where’s that phone book? Acme Visible Records, Inc., West Allview Drive, Crozet, Virginia.

ACME VISIBLE

view additional pages

The $149.95 personal computer.

Introducing the Sinclair ZX81 If you’re ever going to buy a personal computer, now is the time to do it.

The new Sinclair ZX81 is the most powerful, yet easy-to-use computer ever offered for anywhere near-the price: only $149.95* completely assembled.

Don’t let the price fool you. The ZX81 has just about everything you could ask for in a personal computer.

A breakthrough in personal computers The ZX81 is a major advance over the original Sinclair ZX80—the world’s largest selling personal computer and the first for under $200.

In fact, the ZX81′s new 8K Extended BASIC offers features found only on computers costing two or three times as much.

Just look at what you get:
* Continuous display, including moving graphics.

* Multi-dimensional string and numerical arrays.

* Mathematical and scientific functions accurate to 8 decimal places.

* Unique one-touch entry of key words like PRINT, RUN and LIST.

* Automatic syntax error detection and easy editing.

* Randomize function useful for both games and serious applications.

* Built-in interface for ZX Printer.

* 1K of memory expandable to 16K.

The ZX81 is also very convenient to use. It hooks up to any television set to produce a clear 32-column by 24-line display. And you can use a regular cassette recorder to store and recall programs by name.

If you already own a ZX80 The 8K Extended BASIC chip used in the ZX81 is available as a plug-in replacement for your ZX80 for only $39.95, plus shipping and handling—complete with new key board overlay and the ZX81 manual.

So in just a few minutes, with no special skills or tools required, you can upgrade your ZX80 to have all the powerful features of the ZX81. (You’ll have everything except continuous display, but you can still use the PAUSE and SCROLL commands to get moving graphics.) With the 8K BASIC chip, your ZX80 will also be equipped to use the ZX Printer and Sinclair software.

Order at no risk**.

We’ll give you 10 days to try out the ZX81. If you’re not completely satisfied, just return it to Sinclair Research and we’ll give you a full refund.

And if you have a problem with your ZX81, send it to Sinclair Research within 90 days and we’ll repair or replace it at no charge.

NEW SOFTWARE: Sinclair has published pre-recorded programs on cassettes for your ZX81, or ZX80 with 8K BASIC. We’re constantly coming out with new programs, so we’ll send you our latest software catalog with your computer.

ZX PRINTER: The Sinclair ZX Printer will work with your ZX81, or ZX80 with 8K BASIC. It will be available in the near future and will cost less than $100.

16K MEMORY MODULE: Like any powerful, full fledged computer, the ZX81 is expandable. Sinclair’s 16K memory module plugs right onto the back of your ZX81 (or ZX80, with or without 8K BASIC). Cost is $99.95. plus shipping and handling.

ZX81 MANUAL: The ZX81 comes with a comprehensive 164-page programming guide and operating manual designed for both beginners and experienced computer users. A 510.95 value, it’s yours free with the ZX81.

The $99.95 personal computer.

Introducing the ZX81 kit If you really want to save money, and you enjoy building electronic kits, you can order the ZX81 in kit form for the incredible price of just $99.95* It’s the same, full-featured computer, only you put it together yourself. We’ll send complete, easy-to-follow instructions on how you can assemble your ZX81 in just a few hours. All you have to supply is the soldering iron* How to order.

Sinclair Research is the world’s largest manufacturer of personal computers.

The ZX81 represents the latest technology in microelectronics, and it picks up right where the ZX80 left off. Thousands are selling every week.

We urge you to place your order for the new ZX81 today. The sooner you order, the sooner you can start enjoying your own computer.

To order, simply call our toll free number, and use your MasterCard or VISA. To order by mail, please use the coupon. And send your check or money order. We regret that we cannot accept purchase orders or C.O.D’s.

CALL 800-543-3000. Ask for operator #509. In Ohio call 800-582-1364. In Canada call 513-729-4300. Ask for operator #509. Phones open 24 hours a day, 7 days a week. Have your MasterCard or VISA ready. These numbers are for orders only. For information, you must write to Sinclair Research Ltd., 2 Sinclair Plaza, Nashua, NH 03061.

Sinclair

*Plus shipping and handling. Price includes connectors for TV and cassette, AC adaptor, and FREE manual.
**Does not apply to ZX81 kits

The only complete document proofreading system: GRAMMATiK + PROOF READER = NO ERRORS

GRAMMATiK – Beyond Spelling Checking
PROOF READER – The Aspen Software Company Spelling Checker

For CP/M®, TRS-80®, and 8086/8088 Word Processors.

Don’t settle for a partial proofreading program. There are other spelling checkers available, but only Aspen Software offers a complete document proofreading system.

The Aspen Software Company spelling checker is Proofreader. We’ve been improving Proofreader since it was first released in early 1981. The newest version is fully menu driven, and so simple to use thai you probably won’t even need the comprehensive user’s manual that’s included. It has a large 38,000 word fully expandable dictionary, and can check even your largest documents in under four minutes (even faster on most systems). Proofreader’s dictionary is almost twice as large as some, so you won’t have to spend so much time adding new words. And Proofreader is totally accurate. Proofreader looks up every word in its dictionary and does not artificially extend its dictionary size with less accurate prefix and suffix analysis like some others. (Beware: Most checkers with “vocabularies” of 50,000 or more words use this method and can miss some misspelled words!) The entire 38.000 word dictionary takes only 100,000 bytes of disk space.

Proofreader does not simply mark the errors in your document like some checkers, but allows you to make corrections interactively. with the full context of each unknown word displayed. You can correct mistakes, leave unknown words alone, or add words to the dictionary. Unlike most spelling checkers, you also have complete interactive access to the dictionary while correcting. so you won’t need to keep a separate dictionary to look up words manually. Each correction is automatically double checked in the dictionary, and with a single command, Proofreader can almost always show you the correct spelling of a word. (Interactive correction optional on TRS-80 Model I/III, included on all other versions. TRS-80 versions do not support interactive dictionary look up.)

Spelling checking alone is not enough! Aspen Software’s Grammatik goes beyond simple spelling checking. No one else has anything like it. First. Grammatik will check your document

for common typos (such as doubled words: “the the”), and punctuation and capitalization errors (e.g., “STicky shift key”). It also checks for poor writing style using a dictionary of over 500 misused phrases as defined in many writer’s style manuals. Grammatik classifies each error it finds, marks the errors for easy correction with your word processor, and provides suggestions for correcting the problem. The phrase dictionary can easily be expanded to include checking for esoteric jargon or your own personal pet peeves. Grammatik also collects other information that can help you judge the style of the document, and can produce a profile of word usage.

Grammatik is receiving rave reviews from both critics and users. Bob Louden in InfoWorld (12/7/81): “If you use a word processor and a spelling checker, then you should investigate the unique capabilities of this program. Grammatik is a surprisingly fast and easy tool for analyzing writing style and punctuation.” Eric Balkan in The Computer Consultant: “I’m impressed with the imagination that went into this product.” Many users call or write to tell us how much they like Grammatik. Some typical remarks: “Great!”, “Thanks for making my life easier.”, “I’m not just happy. I’m ecstatic!”. Grammatik has also been selected as an officially approved Osborne Computer software package and will soon be appearing at Osborne dealers.

Only Proofreader and Grammatik can provide you with complete document proofreading, and together cost less than some spelling checkers alone. Proofreader and Grammatik have been designed to work with almost any CP/M, TRS-80. or 8086/8088 based word processor. While they have been designed to work together, they are available separately.

Aspen Software also has its own full featured word processor called Writer’s Companion for all these systems. One of the best implementations of Ratfor (Rational Fortran) is available, too, along with an automatic Ratfor pretty printer. Please call or write for more details about these products.

-CP/M versions require CP/M version 2 or later and at least 48K of RAM Standard 8″ single density. Northstar, Osborne 1. Omikron. and Apple formats available directly from Aspen Software. These and most other formats also available from Digital Marketing. Some CP’ M systems with limited disk capacity supplied with 28.000 word. 65.000 byte dictionary. Proofreader-$129.00. Grammatik $150.00. Both – $250.00.

Manuals only -18.00 each. $15.00 for both.

TRS-80 Model I/III versions require only one disk drive and 32K of RAM, and are compatible with all TRS-80 word processors and operating systems. Model II version requires 64K and 1 drive. TRS-80 Model I/III: Proofreader $59.00. Interactive correction option $30.00. Grammatik – $59.00 All $139.00 TRS-80 Model II: Proofreader – $99.00. Grammatik $99.00, Both -$179.00.

-The 8086 8088 version will run under MS/DOS or equivalent (including the IBM Personal Computer DOS). MS/DOS formats available include standard single density 8″ and IBM PC 5.25″. CP/M-86 versions scheduled for Summer 1982 availability. Proofreader $129.00. Grammatik-$150.00. Both – $250.00.

IMPORTANT ORDERING INFORMATION: You MUST specify computer model, operating system, memory size, and format and number of disk drives when ordering either software or manuals alone. Please include your phone number. All U.S.. Canada, and Mexico orders include first class shipping in price. Overseas please add $5.00. We accept cash, check, money order. VISA and Master Card. Sony, no UPS or COD service available. Purchase orders accepted from educational institutions and nationally recognized corporations only. Cost of manual only orders can be credited to final purchase. NM residents add 4% sales tax.

Aspen Software products distributed exclusively by Aspen Software Company and Digital Marketing. Dealer and OEM inquiries welcome. Trademarks: CP/M: Digital Research; TRS-80: Tandy Corp.; MS-DOS: Microsoft; Proofreader, Grammatik: Aspen Software Co.

ASPEN SOFTWARE COMPANY
P.O. Box 339 – A Tijeras, NM 87059 (505)281-1634

Univac at General Electric (Appliance Park)

Offering the unique advantage of electronic speed coupled with unrivalled accuracy, it’s no wonder that the General Electric Company installed the Remington Rand Univac at their modern Appliance Park, in Louisville, Kentucky.

Univac makes possible new insight into all phases of G-E’s major appliance operations. Management decisions can be made with a degree of precision never before obtainable. And G-E also uses Univac to process—automatically—payroll accounting, material control, budget analysis, and sales statistical analysis.

Leading companies throughout the country have learned that Univac has become synonymous with enlightened management. And Univac savings more than justify its use for electronic control of management problems. Find out how typical users have put Univac to work in virtually all types of commercial data-processing. We’ll be happy to send EL135—an informative, 24-page, 4-color book on the Univac System — to business executives requesting it on their company letterhead.

Send your requests to Remington Rand, Room 1853, 315 Fourth Avenue, New York 10, New York.

Remington Rand Univac DIVISION OF SPERRY RAND CORPORATION

Makers of: Univac I * Univac II * Univac Scientific * Univac File-Computer * Univac 60 * Univac 120 * Univac High-Speed Printer

Programmable calculator features accessory ports

A new hand-held calculator I’ve been trying has features—plug-in peripheral slots, scrolling alphanumeric display, “musical” beep-boop sounds— that are familiar to users of typewriter-size personal computers [PS, Nov. '79]. But in a pocket programmable, Hewlett-Packard’s new 41C, these features and others add up to exceptional versatility.

The bare-bones $295 HP-41C has four small sockets behind its liquid-crystal display. These ports accept a variety of plug-ins. Four $45 memory modules, for example, boost the calculator’s capacity from 400 to some 2000 stored program lines. (Lines correspond to data or functions you have to enter each time you solve problems on nonprogrammable models.) You can also plug in combinations of memory and application modules (business, math, engineering, etc.), plus modules customized with your own programs.

But the ports accept more than modules. HP’s $350 portable printer can plug into a port. The thermal printer spews out a variety of alphanumerics, symbols, and high-resolution graphics on heat-sensitive paper. You can also create characters or symbols.

A compact $195 magnetic-card reader plugs into a port, lengthening the HP-41C by about two inches. The reader accepts and stores programs on tiny cards that also work in other current HP models, and some 9000 contributed programs are available.

Often, user-developed programs must be initially entered one keystroke at a time from a program list. But now HP can print user programs in its library as bar codes, similar to those on food packages. Plug in a small optical-wand accessory and you can quickly read programs into memory by sweeping the wand over the narrow black-and-white stripes.

One valuable feature is a continuous memory that retains programs even after you switch the 41C off. The battery (four N cells) lasts nine to 12 months. For example, I stored a program called “Song” by first pushing the alpha key and tapping in this title (the calculator can display letters in the alpha mode). Now, anytime I call this program from memory, the 41C produces 10 audible tones, beeping up and down the pitch scale accompanied by a tiny moving black arrow.

Both tones and alphanumeric messages can prompt you to enter data at points in a program. Short-sentence prompts, over 12 characters, scroll across the display like a billboard advertisement. This alphanumeric capability also calls some of the 130-odd functions not labeled on keys. Entering fact in the alpha mode, for example, calculates the factorial of any number under 69. Errors and messages are displayed as words.

The HP-41C is so programmable that, in the user mode, you can even shift the location of key functions around. A blank clip-in overlay for the keyboard can be labeled with decals identifying reassigned or new key functions. Plug-in application modules include keyboard overlays to identify special-function keys.

A glance through the math-module manual shows how powerful hand-helds have become. Programs include calculus, Fourier series, and (with three memory modules) solutions of a 14-by-14 matrix.— John Free

Scientific calculator has 24-digit rolling display

A rolling 24-digit liquid-crystal display (LCD) on Sharp’s new EL-5100 calculator, plus other unusual features, makes it the easiest-to-use scientific I’ve tried. Tap in a problem and you’ll see each element displayed as though you were typing.

Formula statements such as sinh^-1 (arc hyperbolic sine), requiring five readouts, fill up the black-and-gold display quickly. No problem. Long for- mulas or chain calculations appear to scroll left. There’s room in memory for 80 formula steps (sinh^-1 only counts as one step).

Enter a formula in the Algebraic Expression Reserve (aer) mode and months later it’ll still be stored. In fact, you can store five separate formulas in continuous memory (80 steps total) and call each back at the touch of a key.

Push the playback (pb) key to display a formula and a blinking rectangular cursor appears. As it blinks, you can see that each readout is a five-by-seven array of tiny dots. Two arrow keys let you shift the cursor rapidly about, making corrections, additions, or deletions. This editing feature is exceptionally useful.

Ten data memories, accessed through keys a to j, store variables, constants, or results. After jotting down a list of letter equivalents to formula elements, you can alter any memory directly and note the effect on a solution. The 5100 can also prompt you through entry of each variable (a=?, b= ?, etc.) with a computer-like blinking cursor.—J. F.

view additional pages

How do you decide who gets priority on your computer?

new 6000 SERIES Systems make “priority” a thing of the past

YOU CAN MAKE EVERYBODY “FIRST IN LINE” — because the new CONTROL DATA® Series 6000 Systems do things differently than any other computers available today. Their massive memory and incredible speed allow simultaneous access by a number of different users with different programs.

Each of these Series 6000 Systems is literally eleven computers in one — “concurrent parallel” — able to operate separately or in concert as needed. They handle eleven programs at once . . . along with almost unlimited peripheral equipment. And operate fantastically fast; ten to a thousand times more rapidly than other present systems.

There now are three sizes of Series 6000 Systems to choose from … the 6400, 6600 and 6800. You might use one to — • Provide computing for your entire corporate complex, streamlining research, production, inventory control, accounting, market appraisal, order processing and management. Handle all the operations concurrently, as if each division had its own system.

• Revolutionize order handling for retail chains with geographically separated outlets. Supply on-the-spot confirmation of goods available in scattered warehouses. Issue restocking orders. Check credit. Do accounting on an automated, continuing basis. All at the same time.

• Correlate data of enormous proportions and interlocking meaning for engineers and researchers. Study 3-dimensional mathematical models beyond any conceived up to now. Handle all elements of these immense problems simultaneously, in real time.

For complete documentation of the capabilities of Series 6000 systems, write to Dept. B-45, Control Data Corporation, 8100 34th Avenue South, Minneapolis, Minn. 55440.

Cost?

Not higher —but less!

Anyone presently spending $20-thousand a month rental or more should investigate a Series 6000 System. One prominent manufacturer now planning a 6000 installation anticipates savings from the switchover of roughly $100-thousand a month because of the improved price-performance ratio. And Series 6000 Systems offer you a two-year delivery lead over any alternative computer that can claim to be in the same league … install in less space than many present systems of far less capability.

CONTROL DATA CORPORATION
8100 34th AVENUE SOUTH, MINNEAPOLIS, MINNESOTA 55440

Computer system speaks plain English

Computers may be storehouses of information, but to release it they must be addressed in a special language. Now Philips in Holland has devised a system that understands questions in English, so more people have access to the data bank. This gives an organization’s computer far wider use, and points to exciting future developments for home units.

Queries can be phrased with any choice of words or sentence structure, and the computer understands them through a three-step electronic process. First, the grammatical structure of the sentence and the logical relationships between its parts are established. For this, the primary memory stores an English grammar. In the second step, nouns and verbs are identified by reference to a data bank with a selected vocabulary related to the particular topic. Finally, the question is translated into “data-base language.” Search instructions reach the main data bank in this form, and the answer comes up on the visual display unit. If the reply is ambiguous, the computer will search for another interpretation, and the whole process is repeated.

The system, called PHILQA (Philips Question Answering), is still at the early development stage. For the first experimental tests, the data base has fictitious information about computers in use in Europe. So far, the answers are very brief, and limited to “yes” or “no,” a number, or a list of names and addresses. The system can also show favor for one of several possible answers, or indicate that no answer is available. It can’t refer back to a previous answer. —David Scott

“The computer,” says Atari’s Steve Jobs, “is one of the pinnacles of Western rational thought. They bring together physics, electronics, chemistry and mathematics; they bring logic, and philosophy, information theory, all that. And the people working on these computers possess a passion about the discovery and creation of something. It’s a passion that I have only seen matched in people pursuing what they consider to be the truth of their existence. It’s the same purity of spirit I have experienced in monks.”

Also check out his thoughts in this interview with Byte Magazine done when the Macintosh was introduced.

The hip young heroes of Atari, for instance, are convinced that they stand on the very brink of evolutionary breakthrough. The development of the video games is seen as roughly equivalent to mankind’s slow crawl from the primal broth of creation. Any day now, it seems, homo sapiens will once more be toweling himself down on the fresh dunes of tomorrow. “The computer,” says Atari’s Steve Jobs, “is one of the pinnacles of Western rational thought. They bring together physics, electronics, chemistry and mathematics; they bring logic, and philosophy, information theory, all that. And the people working on these computers possess a passion about the discovery and creation of something. It’s a passion that I have only seen matched in people pursuing what they consider to be the truth of their existence. It’s the same purity of spirit I have experienced in monks.” So perhaps the foul-mouthed, grimacing youths of the arcades aren’t just improving their geometrical and spatial awareness: what they’re really doing is searching for the meaning of life.

HEATHKIT H8

This 8-bit machine, by itself is as versatile as a lot of systems that include peripherals

Skeptical? For starters, because of its unique design the H8 is the only machine in its price class that offers full system integration yet, with just 4K of optional memory and using only its “intelligent” front panel for I/O, may be operated completely without peripherals! In addition, by using the features of its built-in Pam-8 ROM panel control program, the H8 actually allows you to dig in and examine machine level circuitry.

Responding to simple instructions the “intelligent” panel displays memory and register contents, lets you inspect and alter them even during operation. And for greater understanding, the front panel permits you to execute programs a single instruction at a time. The result is a powerful, flexible learning tool that actually lets you “see” and confirm each detail of H8′s inner workings.

If you need further evidence, consider the fact that H8′s system orientation allows you an almost unlimited opportunity for growth.

Memory is fully expandable, the 8080A CPU extremely versatile, and with the addition of high speed serial and parallel interfacing you gain the added flexibility of I/O operation with tape, CRT consoles, paper tape reader/punches, and now our new floppy disk systems! The H8 offers superior documentation including complete step-by-step assembly and operation manuals, is backed by 54 years of Heath reliability, and comes complete with BASIC, assembler, editor, and debug software — others charge over $60 for!

H8, simplicity for the beginner, sophistication for the expert and at $375* just right for you.

HEATHKIT COMPUTERS System Engineered for Personal Computing

view additional pages

VIC-20 – Commodore’s Entry in the Small Computer Arena

by David D. Busch

If first impressions stick, the Vic-20 microcomputer by Commodore (King of Prussia, PA) will lodge itself in the mind of any potential purchaser. The 6502 microprocessor-based computer just doesn’t look like a $299 machine.

In fact, when I demonstrate the unit to those unfamiliar with it, I always save the price for last. This ploy is especially effective if the potential user already has some familiarity with other microcomputers and their prices.

First, I demonstrate the full-stroke, typewriter-style keyboard, which features four special function keys and a control key. The Pet Basic is identical to that used in higher-priced Commodore machines and comparable to Applesoft or Radio Shack’s model III Basic. Lowercase characters are built-in, but the user can define any character set by using an 8-by-8 dot matrix. There are high resolution graphics (every dot on the screen may be addressed)—in eight colors. Need sound? Four voices are built in, and adding a joystick is as simple as plugging a stock Atari unit into a socket.

I then mention the price. “What’s the catch?” I am asked immediately. In truth, there are several. It is not really possible to do anything useful with the system for $299—not even save or load programs. It is necessary to add $75 for the Commodore Datasette cassette recorder. Ordinary tape recorders, which produce sine wave output, cannot be substituted for the square-wave Commodore unit.

Because the unit comes with 5K bytes of RAM and only about 3.5K bytes are available for user programs, it is necessary to purchase additional memory in order to use all of the features to their fullest capacity. For example, in the stock Vic-20, only a small portion of the screen can be manipulated in hi-res graphics. Defining a complete new character set leaves only a small amount of memory for programs. These tools can be best used when an additional $300 investment is made to expand RAM to its 32K-byte limit.

The final catch Is that the display is only 22 columns wide. This may also be remedied by add-ons and user programs. However, at a time when many micro users are complaining that 64-character screens should be 80 columns wide, a 22-character display seems less than adequate.

My feeling is that Vic-20 purchasers should appreciate the value they get for $299, but consider the unit for what it really is—a competent $600-$700 color microcomputer.

The company can easily be forgiven for the attention given to the keyboard while offering a less-than-perfect screen display. The screen display is relatively easy to modify with add-on components. Given the wide range of televisions likely to be used as monitors for low-cost micros and narrow bandwidths, Commodore opted to provide a sharp display that can be upgraded to suit the higher resolution of composite video monitors.

As a writer capable of 60-word-per-minute bursts, it was the keyboard that sold me. It’s a solid, professional-feeling unit that can accept input as quickly as any user can type. There is no keyboard bounce, and most keys have been laid out in a logical manner.

The biggest drawback is the placement of the RETURN (ENTER) key. It is at the far right of the home row, four keys removed from the L. I am most accustomed to the following layout: “J, K, L, semicolon, ENTER,” although one of the computers I use follows a “J, K, L semicolon, quotation mark, ENTER” format, both standard, typewriter-style layouts. The Vic-20 uses “J, K, L, colon, semicolon, equals-sign, RETURN.” As a result, the tendency is to hit the equals-sign key instead of RETURN.

It is relatively easy to learn this unusual layout and by the time I had written two or three programs, I rarely hit ” = ” by mistake. All other keys have logical or unobtrusive placement.

One welcome feature is the addition of four special function keys. These are marked F1, and so on, and by hitting SHIFT, are used to produce eight separate function keys. They are defined as CHR$(133) through CHR$(140), and are accessible to any Basic program. Business users and game enthusiasts will find special applications for these keys. For example, a menu may be designed instructing the operator to hit F1-F8 to invoke particular subroutines. Then, lines such as IF A$ = CHR$(133) GOTO 100 may be used. An alternative might be: ON VAL(A$) – 133 GOSUB 100,200, 300,400,500,600,700,800.

A large number of characters are available from the 61-key board. The machine powers up in Graphics mode and typing normally produces the uppercase characters and numbers shown on the keytops.

However, each key may also produce one of two graphics characters while in the Graphics mode. Two keys are located on the lower left side of the keyboard; one of two SHIFT keys is to the right, while a special key labeled with the Commodore logo is on the left. Striking the Commodore key and pressing another key produces the graphic character printed on the left side of the keyfront. Pushing the SHIFT key in tandem with another key produces the graphic character printed on the right side of the second keyfront. (Of course, either SHIFT key may be used, but it is easier for a beginner to stick to the one paired with the Commodore key.) The Vic-20 has a Text mode, and by pressing SHIFT and the Commodore key simultaneously, Text is invoked. The keyboard then functions as an ordinary typewriter with upper and lower case letters. Even so, the graphics on the left sides of the keyfronts (those most likely to be used in business applications) are still available by hitting the Commodore key.

That’s not all. Any of the characters available may be printed in any of eight colors, normally or reversed, by using a Control key. The Vic-20 prints its characters in blue until a color control key is pressed. The color keys are on the top row of the keyboard. CTRL-1 produces black characters, CTRL-2, white, and so on through red, cyan, purple, green, blue and yellow. Reverse On and Reverse Off are also summoned by using the control key. Once one of these has been entered, the display changes to the new parameters for all characters until the next control character is entered.

Vic-20′s Pet Basic does not have any graphics or color commands, such as Line or Fill, but I understand that the Super Expander cartridge will provide these capabilities. Graphics characters may be incorporated in programs through time-consuming, but easy-to-use PRINT statements. For example: 100 PRINT “(RVS ON)(CTRL-3) THESE CHARACTERS WILL BE PRINTED IN THE BACKGROUND COLOR ON RED”

110 PRINT “(RVS OFF) THESE CHARACTERS WILL BE PRINTED IN RED, OVER THE BACKGROUND COLOR.”

For the alpha characters, I could have substituted any of the graphics characters, but, of course, could not reproduce them for the printed page. In fact, the (RVS ON) and (CTRL-3) in the above lines actually appear on the Vic screen, replaced by graphics symbols that denote the function invoked. All screen controls are indicated in this manner—a reverse “Q” means cursor down, while a heart indicates clear screen. I found these symbols confusing at first. Unless you have a printer with graphics capability (Commodore offers one for the Vic-20 at $399), your program listings will be lacking all the needed information.

I found the screen display to be excellent. The unit connects to any color television through a supplied RF modulator. I tried the unit with seven different brands and had no trouble getting a sharp image. One exception was a 10-year-old television with a dirty tuner mechanism. Fine-tuning the TV, and, sometimes, moving the RF modulator to get rid of a residual moire effect produced an excellent picture. My test for picture quality involved the use of a high-resolution game, AMOK, a machine language program with a special character set that makes considerably greater demands on resolution than standard alphanumerics or graphics.

POKE generates color, sound, noise On some sets, color balance adjustments had to be made to clearly differentiate the eight available colors. Cyan tended to resemble green on a few units, while red often looked more brown or orange. To avoid constant plugging and unplugging, I have dedicated a 13-in. GE television to permanent Vic duty in my office. Until I receive the Super Expander cartridge and begin working more heavily with hi-res graphics, I see no need to invest in a monitor for the Vic.

The display area on the screen is surrounded by a colored border. Both the border and the screen may be changed to other colors with a single POKE statement, either in command mode or from a program. There are 255 color combinations available.

POKE may also be used to generate sound and noise. Three voices, each with a different three-octave range and a fourth white noise tone generator, are available. The Vic introductory guide includes programs to simulate 20 sounds, including laser beams and birds chirping. Some fine tuning on the television may be necessary to optimize the sound with picture.

Programs are loaded and run from the Pet Datasette cassette recorder. This unit is totally slaved to the CPU, deriving its power and on-off signals from the main unit. At 300 baud, it is a little slow, but I found the capability to use long file names extremely helpful. With Radio Shack cassette files, for example, one may CLOAD or CSAVE a file by single letter name—CLOAD A, for example.

The Vic-20 may be commanded to search for a more-detailed file name—LOAD CHECKBOOK 1982—for example. Or if the file name is not known for certain, a partial specification may be entered. LOAD CHECK would cause the Vic to load CHECKBOOK 1981, CHECKBOOK 1982, or CHECKERS, depending on which it found first. However, the system does tell you what files it has found as it passes through.

I have tried several programs that use the Atari joystick. At present, only a single controller may be used. AMOK, based roughly on the arcade standard Berserk, is a Roger Merrit product. Using the joystick, I was able to make the little man run through the maze with ease, zapping robots along the way. The joystick may be read from user programs through a simple PEEKing routine described by Commodore in its Programmer’s Reference Guide.

Although not tested, light pens and game paddles can also be used with the Vic. I see this machine, because of its low cost, becoming a popular games machine—like the Atari 400—with some serious capabilities as an added attraction.

Those are the current hardware highlights—disk drives, memory expansion and ROM packs. Other enhancements are on the way, but unavailable at press time.

Software base is firm From a software standpoint, the Vic-20 enters the field from an unusually strong position. All standard Pet Basic programs may be loaded and run, as long as ROM-dependent features (such as PEEK and POKE) are not used. Programs written for 40- and 80-column screens will cause material to wrap around the Vic’s 22-character display, but the user can usually remedy this minor annoyance.

Those who already know Pet Basic, or some other Microsoft variation on that theme, will find transition to the Vic painless. TRS-80 owners will be hard-pressed to note any differences at all—GETS in place of INKEY$ and FRE(X) for PRINT MEM are the two notable examples. Even file handling is very similar to Radio Shack’s Disk Basic. Files may be opened and closed, but the Vic-20 allows specification of a device number so that the same syntax may be used for either cassettes or disk data files.

Because of the common language, the Vic-20 should prove to be formidable competition for the Radio Shack Color Computer as a second computer for current model I or III owners.

As a long-time user of line-oriented program text editors, it took me several weeks to become accustomed to the Vic’s screen editing. With line editing, the programmer LISTS a section of the program, types in EDIT 20 (or whatever line is to be changed) and then makes changes. The Vic’s screen editor works in a very different way.

Any program line that appears on the screen may be edited by moving the cursor to that line, overtyping the text, or making insertions or deletions by hitting an appropriate key. Pressing RETURN makes the changes permanent. One may simply renumber a line by typing over the line number (the old version of the line remains and should be deleted).

Beginners to computing will find the guidebook, Personal Computing on the Vic-20, to be a clear, concise and helpful introduction. However, I have seen better Basic courses. The Commodore guidebook is more feature-specific and helps the new user discover the capabilities of the Vic-20 rather than serving as a thorough grounding in Basic programming.

Those already familiar with some dialect of Basic who choose the Vic-20 will find the introductory guide somewhat frustrating. It is difficult to sort through the elementary material to find the “meat” to help you make the switch. The guide is not at all detailed in helpful programming tricks. For example, I could find no reference to using the special function keys. I had to experiment and finally used GET$ to see what ASCII value was returned when the user pressed one of the special keys.

Serious programmers must make the purchase of the Vic-20 Programmer’s Reference Guide their number one priority. This volume not only contains all the material missing from the user’s guide, but has invaluable data I never expected to see. The reference guide contains techniques—with examples—that left me more impressed than ever with Vic’s capabilities.

Consider generating special character sets. There is a register in RAM that controls where the Vic gets its character set information. This location may be changed by hitting the shift Commodore Key, or by POKING to toggle back and forth between two ROM addresses which contain the Uppercase/full graphics or Uppercase-lowercase/partial graphics characters. These character sets may also be reversed. Because the characters are stored in ROM, it is obviously impossible for the user to change them. However, by POKING the RAM register which supplies the ROM addresses with a new location in RAM, it is possible to have the system derive its character information from a user-supplied table. Then, when letter A is printed, instead of A, the screen will display whatever character has been created by the user.

Or, most of the existing character set may be copied over from ROM to the RAM table, so that most keys still produce their ordinary output while others generate new characters. Each complete character set takes up 2K bytes of memory, so not much room is left for user programs with an unexpanded Vic.

The method for designing characters is simple to understand. Each is built on an 8-by-8 matrix, so that eight bytes (each containing eight bits) are needed. It’s easier to visualize this capability:
Bit 1 Bit 2 Bit 3 Bit 4 Bit 5 Bit 6 Bit 7 Bit 8
Byte 1 0 0 0 1 1 0 0 0
Byte 2 0 0 1 0 0 1 0 0
Byte 3 0 1 0 0 0 0 1 0
Byte 4 0 1 1 1 1 1 1 0
Byte 5 0 1 0 0 0 0 1 0
Byte 6 0 1 0 0 0 0 1 0
Byte 7 0 1 0 0 0 0 1 0
Byte 8 0 0 0 0 0 0 0 0

The above eight bytes, translated from binary and POKED to the desired location in the user’s character set, will produce the uppercase letter A. By placing X’s and O’s in a similar matrix, users can custom-design special languages, scientific symbols or flying saucers for games.

Machine language programmers who don’t like rewriting their codes to be compatible with every new ROM will appreciate Commodore’s dedication to the KERNAL concept. KERNAL, Vic’s operating system, contains a standardized jump table to the input, output and memory-management routines in the system. Should Commodore introduce new ROMs for the Vic (as they have done for the Pet), the KERNAL jump table will be changed to match new ROM locations for routines. Serving as a sort of forwarding address, it will allow programmers to address the ROM through the table, confident that JSRs will end up where they are supposed to—even if the ROM has been changed.

The Vic-20 will inevitably be compared with other computers in the low price range. I’ve owned a Sinclair ZX81, and have investigated both the TRS-80 Color Computer and the Atari 400, and can make some comparisons.

The most significant difference among the four is the keyboard. As a touch typist, I found the tiny Sinclair membrane keyboard almost impossible to use for long periods. The single key entry of most Basic keywords was of no help because I can usually type PRINT faster than I can hunt for the correct key on the ZX81′s tiny surface.

The Atari 400 keyboard is only slightly better. It is roughly the size of an actual typewriter keyboard, so the fingers may be positioned for touch typing. However, the flat keys provide no feedback, and it is possible to do only slow, methodical touch typing with it. The Color Computer’s calculator keys are better, but still not up to the full keyboard feel of the Vic.

Costs vary with power and equipment On the basis of price alone, the Sinclair is quite powerful for $150 ($99 as a kit), but its 32-character by 16-line CRT display is black and white only, and resolution may be spotty on a variety of televisions. Of course, the unit comes with just 1K byte of memory, and expanding to 16K bytes raises its price $100. That is still less expensive than the $299 Vic, which actually costs closer to $400 with the required cassette recorder included.

That makes the Vic similar in price to the $399 Color Computer, which may be used with a tape recorder in a pinch. The Atari, also $400, is supplied with 16K bytes of memory, which may make it a better deal if all factors were equal.

The Atari will display 24 by 40 characters, to the Color Computer’s 32 by 16, and Vic’s 22 by 23 (without expansion add-ons).

The ZX81 is now limited to 16K-byte memory, but the other three systems can be expanded to 32K bytes with memory boards supplied by either their manufacturers or outside suppliers. The three color computers can also use ROM packs.

As far as other peripherals go, the Color Computer has an early lead, because its disk drives are now available. Outside suppliers have also introduced accessories for the Color Computer’s expansion bus, which will make that unit very flexible should it remain in the Radio Shack line for a number of years.

Disk drives for the Vic have yet to be introduced, although one is promised. A printer is already available, and third-party suppliers offer RS-232 interfaces and other accessories.

As with any computer purchase, the final decision should be partially based on end-user requirements. On business trips, I carry a complete ZX81 kit in one pocket of my camera bag. For that application, small size and portability were paramount. Now I find the Vic-20 has its advantages. It is a low-cost color computer with a Basic that took me no time to learn. Because it has built-in lowercase and is portable, I may find some future word processing-on-the-road uses for it.

Others may prefer the Atari because an excellent selection of games is already available and the Color Computer might be the choice for some because of widespread Radio Shack support.

To date, support for the Vic from Commodore has been slow. Promised accessories and peripherals were two to three months late. My local dealer noted that he sold 50 units during the Christmas season, but would have been able to move 200 or more if more software and hardware had been available. The thoroughness of the Programmer’s Reference Manual is a good sign, and could indicate an exciting future for the Vic-20.

Also, Philip Elmer-De Witt still writes about technology.

And apparently Seymour Cray was so bad-ass he played Minecraft for real.

view additional pages

Fast and Smart – Designers race to build the supercomputers of the future

The computer at the University of Illinois is simulating something that no one saw: the evolution of the universe in the aftermath of the Big Bang. Re-creating conditions that may have prevailed billions of years ago, the computer reveals on a remote screen how massive clouds of subatomic particles, tugged by their own gravity, might have coalesced into filaments and flattened disks. The vivid reds, greens and blues of the shapes are not merely decorative but represent the various densities of the first large structures as they emerged from primordial chaos in the near vacuum of space.

At the Massachusetts Institute of Technology, another computer is struggling to learn what any three-year-old child already knows: the difference between a cup and a saucer. What the youngster sees at a glance, the computer must be taught, painstakingly, one step at a time. First it must comprehend the concept of an object, a physical thing distinguished from the space around it by edges and surfaces. Then it must grasp the essential attributes of cupness: the handle, the leakproof central cavity, the stable base. Finally, it must deal with the exceptions, like the foam-plastic cup whose heat-insulating properties are so good that it does not need a handle.

These experiments illustrate the paradox at the heart of today’s computer science. The most powerful computing machines—giant number crunchers pos- sessed of speed and storage capacities beyond human comprehension—are essentially dumb brutes with no more intellectual depth than a light bulb. At the other extreme are computers that have begun to exhibit the first glimmers of humanlike reasoning, but only within the confines of narrowly defined tasks.

For 40 years scientists have labored to make headway at these two frontiers of computer research. One group, working with the lightning-fast machines known as supercomputers, is always pushing for more raw power, more blazing speed. The other group, writing programs that show the rudiments of artificial intelligence, explores the mysteries of human thought. Each of these two grand scientific enterprises, backed by billions of research dollars and blessed with some of the century’s best minds, has proceeded as if the other did not exist.

But there are signs that the two broad avenues of computer research may be starting to converge, that today’s most advanced machines may someday evolve into electronic brains that are not just incredibly fast but smart as well. The quest has been taken up by almost every major nation. And no wonder: the potential rewards—in industrial productivity, scientific research and national security—are staggering. Grown men glow with childlike excitement when they describe robots that will see their way around a factory, typewriters that will take dictation, defense systems that will make the world safe from nuclear arms.

The two fields of computer research are at different stages in their life cycles. Artificial intelligence is just getting started: the first commercial projects appeared less than five years ago, and are now finding widespread application (see followingstory). The supercomputer manufacturers.

on the other hand, having supplied highspeed processors to government labs and intelligence agencies for a quarter-century, are now experiencing a growth so explosive that it has taken even the most optimistic industry leaders by surprise. Sales of the machines, which cost $5 million to $25 million each, have increased 25% a year or more over the past decade, and in 1988 will pass the $1 billion-a-year mark for the first time.

Some 300 supercomputers now work at tasks as diverse as ferreting out oil deposits. analyzing muscle structures and creating special effects for Hollywood films. With the spread of supercomputer networks, high-speed computing power is available to anyone with a personal computer and a telephone hookup. “The world will never be the same,” says Doyle Knight, director of the John von Neumann National Computer Center in Princeton, N.J. “Soon every industry, every science, every walk of life will in some way be touched by supercomputing.”

Speed and power are what distinguish supercomputers from their humbler cousins. In the early days of the industry, speed was measured in thousands of FLOPS, an acronym for floating-point operations per second, in which the decimal point is moved in very large and small numbers. Today’s largest machines are measured in gigaFLOPS, or billions of operations a second. Tomorrow’s will be measured in teraFLOPS, trillions of operations a second. A single supercomputer going at teraFLOPS speed will have the power of 10 million personal computers working at full throttle.

The most powerful supercomputers are surprisingly small and sleek, some not much bigger than a California hot tub. But looks can be deceiving. Supercomputers often squeeze out the last bit of processing speed by shrinking the distances electrons have .to travel within their wiring. They are tightly packed workhorses that require a whole array of supporting equipment. Some employ full-size mainframe computers just to shuttle programs in and out of their processing units. The machines may be connected, by cable or satellite, to hundreds of remote terminals that can transform raw numerical output into stunning 3-D graphics. They often need industrial-size refrigeration units to keep the rush of electronic signals within them from melting down their circuitry. The thermal output of the University of Minnesota’s supercomputers is used to heat a garage.

For most of the supercomputer era, the market for the most powerful machines has been dominated by one firm, Cray Research of Minneapolis. With 178 of its distinctive C shaped models installed around the world, Cray accounts for 60% of all the supercomputers sold. The closest competitor, located directly across the Mississippi River in St. Paul, is the company from which Cray split off in 1972: Control Data Corp. CDC, which in 1983 created a supercomputer subsidiary called ETA Systems, is holding steady with a 12.7% mar- KET share. Coming up quickly is a trio of Japanese manufacturers—NEC, Hitachi and Fujitsu—that entered the supercomputer race in 1983 and has since captured 23% of the world market.

But this tidy pie chart may soon be upset by the surprise entry of a new player that for the past two decades has been most conspicuous by its absence from the supercomputer market: IBM. In December the largest computer manufacturer (1987 sales: $54.2 billion) announced that it had struck a deal with Steve Chen, one of the foremost supercomputer designers, who jolted the computer world last September by suddenly leaving his post as a vice president at Cray. With financial aid from IBM, Chen has set up his own company to develop a machine 100 times as fast as any currently on the market. “People say that IBM is just dipping its toes into the water,” notes Irving Wladawsky-Berger, an IBM vice president. “We’re in the middle of the ocean.”

IBM has not only taken the plunge but has also put its prestige and enormous resources behind a radical kind of supercomputer that represents a dramatic break from the past. Since World War II, most computers have been designed to do things one step at a time, moving data in and out of a single high-speed processor. The computer Chen is building with IBM’s backing will contain not one but 64 processors, all operating at the same time, in parallel, and thus significantly cutting down computing time. IBM’s decision to support a major parallel-processing supercomputer project is a sign that technology is headed in that direction. Says H.T. Kung, computer scientist at Carnegie-Mellon University: “In one move, IBM legitimized two technologies: super-computing and parallel processing.” AT&T Bell Laboratories is expected to introduce a new parallel-processing computer at the American Physical Society meeting in New Orleans this week.

Cray, IBM and AT&T could be upstaged, however, by a determined gang of innovative computer designers who have already moved beyond 64 processing units to build machines that divide their work among hundreds, even thousands of processors. Last week scientists at Sandia National Laboratories in Albuquerque announced that they have coaxed a 1,024-processor computer into solving several problems more than 1,000 times as fast as a single-processor machine acting alone, an unprecedented speedup that suggests the performance of supercomputers may in the future be related almost directly to the number of processors they employ.

Much supercomputing research is funded by the U.S. Government, whose appetite for high-speed, number-crunching power for both defense and intelligence uses seems boundless. Last year the Pentagon spent hundreds of millions of dollars trying to step up the speed of the fastest machines. One Government project that has a special need for supercomputing power is the national aerospace plane, a high-altitude aircraft intended to carry military and civilian cargo at up to 25 times the speed of sound. Since there are no wind tunnels capable of simulating such blistering airspeeds, the hypersonic plane will have to be tested on supercomputers, ideally on machines many times as powerful as existing models. Presidential Science Adviser William Graham has recommended that Congress appropriate an additional $1.7 billion to support the development of parallel-processing supercomputers that by the mid-1990s could crunch data at teraFLOPS speed.

The military-intelligence connection is nothing new for supercomputer manufacturers. One of the first Crays to come off the assembly line in 1976 was shipped to the Lawrence Livermore National Laboratory, where it made short work of the mind-boggling mathematical equations required to design hydrogen bombs. Another early Cray without doubt was delivered to the National Security Agency in Fort Meade, Md., where it would have been put to work cracking military codes and sorting through the intelligence data that flood into the agency every day.

What is new is the rapidly growing appetite for supercomputer power in the private sector. In a classic case of a technology developed for a few specialized purposes finding application in all sorts of unexpected areas, supercomputing has spread from one industry to another like a benevolent virus. Semiconductor manufacturers use supercomputers to design ways to squeeze more transistors into a square-centimeter chip of silicon. Financial advisers use them to devise investment strategies of dizzying complexity. Biochemists need them to predict which molecules are worth testing as new medicines. Engineers rely on them to design new cars, jet engines, light bulbs, sailboats, refrigerators and artificial limbs.

No one benefits more from supercomputing than research scientists. The National Science Foundation belatedly recognized that fact in 1985, when it committed itself to spending more than $200 million to create supercomputer centers at five selected sites, plus the electronic links to connect the machines to dozens of universities and research labs. Today some 6,000 scientists at more than 200 institutions have access to the NSF centers. This availability has sparked a burst of scientific productivity in fields ranging from mathematics to fluid dynamics. Says Ron Bailey, chief of the Numerical Aerodynamic Simulation program at the NASA Ames Research Center: “Supercomputers are as significant to pioneering research today as calculus was to Newton.”

Supercomputers are giving scientists unprecedented access to hidden worlds both large and small. Using the prodigious power of the Cray at the San Diego Supercomputer Center, Researchers Mark Ellisman and Stephen Young are studying a pair of noodle-like structures in the brains of Alzheimer’s victims that scientists think may be a cause of premature dementia. Northwestern University Professor Arthur Freeman used a Cray-2 to produce a stunning portrait of the atomic structure of a new superconductor that carries an electric current freely at -283° F. The Cray X-MP at the University of Illinois has produced a dazzling array of colorful animations, from the roiling birth of a tornado to the supersonic fountains that spew forth from black holes at the centers of galaxies. Says Nobel Physicist Kenneth Wilson of Cornell University: “An astronomer with a telescope can observe the universe over a period of 50 years. But an astrophysicist with a supercomputer can ‘see’ billions of years into the past and the future.”

Yet for all the miracles supercomputers have made possible, their users are still not satisfied. Computer Architect Neil Lincoln jokes that the real definition of a supercomputer is a machine that is just one generation behind the problems it is asked to solve. Norman Morse, head of computations at Los Alamos National Laboratory, has eleven supercomputers at his disposal but still cannot please his weapons designers and other scientists. Says he: “We already have jobs right now that require a machine 100 times as fast as anything we have.”

The race to build those faster supercomputers is well under way. In dozens of laboratories in the U.S., Europe and Japan, millions of dollars are being spent to support the efforts of hundreds of engineers and scientists, all driven by the dream of building the world’s most powerful computing machine. If any one team can be said to have the head start, it is the small, tightly knit group of technicians working in an industrial park in Chippewa Falls, Wis., where Cray Research has its most important laboratories.

Chippewa Falls (pop. 13,000) enjoys a local reputation for its Leinenkugel’s beer and Chippewa Springs water. But it is known around the world as the home of one of the most influential and enigmatic figures in computer science: Seymour Cray. A shy, solitary engineer who rarely gives press interviews, Cray, 62, is to supercomputers what Edison was to light bulbs or Bell to the telephone. First as a co-founder of Control Data, then for his own firm, Cray has designed an extraordinary series of high-performance machines, including the CDC 1604 (1960), CDC 6600 (1964), CDC 7600 (1969), Cray-1 (1976) and Cray-2 (1985), each of which could at the time lay claim to being the world’s most powerful computer.

In 1981 Cray stepped down as chairman of the company and became a “consultant,” but that only gave him more time to focus on computer design. He is now completing plans for his next machine, the Cray-3 (due to be released in 1989), and is soon expected to focus on its successor, the Cray-4, with a single-mindedness that is legendary. “Seymour has the ability to concentrate on his work to the wholesale exclusion of everything else,” says James Thornton, a former engineering colleague at CDC. “He captures the universe of what he’s going to design inside his head, and there he stays until he’s through.”

Technologically, Cray shows no signs of losing his innovative touch. The Cray-3 is expected to be the first commercial computer to use chips made of gallium arsenide as well as the usual silicon. Electrons travel up to ten times as fast through gallium arsenide, and although the material is more difficult and costly to work with, Cray has determined that the gain in speed will justify the added expense. Recognizing the growing importance of parallel processing, Cray is planning to give his most advanced model 64 processors, instead of the four in the Cray-2 and the 16 that will go into the Cray-3. Yet Cray is careful not to move too far, too fast. “The concept of stride is very important in developing computers,” Cray told a group of customers last fall. “If you take a stride that is too large, you get bruised. If you take a step in one dimension, you better be careful about taking a step in another, or the step may get too long.”

Those comments were taken to be a pointed reference to the work of Steve Chen, 44, who left Cray Research abruptly when the company refused to go along with his plan to build an ambitious new machine. By the time he walked out the door, Chen was already a star in the supercomputer field. Born in China, he grew up in Taiwan, moved to the U.S., studied electrical engineering at Villa-nova and got his doctorate at the University of Illinois in Champaign/Urbana. When he came to Cray Research in 1979, the company’s officers thought they had found Cray’s successor: someone as brilliant and dedicated as the master himself. Chen certainly aspired to be in the same class as Cray. Says he, with characteristic modesty: “There are only a few people crazy enough to do this all the time.”

Chen quickly proved himself by reconfiguring the Cray-1 as a two-processor machine. The resulting computer, the Cray X-MP, became the best-selling supercomputer of all time, with more than 120 installed. Chen also designed the newly introduced Cray Y-MP, which the company hopes will match the commercial success of the X-MP. But Chen’s drive to build ever more powerful computers brought him into conflict with Seymour Cray. The problem, according to Gary Smaby, a vice president at the Piper, Jaffray & Hop-wood investment firm in Minneapolis, was not jealousy or a clash of personalities but contrasting technological styles. Cray’s genius has been to get the most out of existing technology with a tight budget and skeleton staff. Chen took a “team approach,” hiring a staff of 200 and encouraging them to push the state of the art wherever possible. In the eyes of management, Chen’s proposed machine, the Cray MP, would have involved risk on five different technological fronts, including the limited use of fiber-optic cables to send some streams of data with beams of light rather than electrons. When projected costs hit $100 million, more than double the original budget, Cray Chairman John Rollwagen backed away and canceled the project outright, forcing Chen’s resignation.

About 45 members of his research team at Cray defected along with Chen and set up shop twelve miles away in Eau Claire, Wis. Within three months Chen lined up the financial commitment from IBM, estimated at $10 million to $45 million. “We know what it takes to nurture visionaries,” says IBM’s Wladawsky-Berger. “We want Chen to swing for the fences.” And that is what he intends to do. Says Chen: “Five years from now we should be at 100 billion gigaFLOPS. A problem that takes three months to do now, we want to do in a day.”

IBM is not relying solely on Chen, however. As the supercomputer market reaches the magic $ 1 billion-a-year figure that has traditionally been the company’s threshold of interest, IBM has at least six different supercomputer efforts under way, although some are primarily research projects. One experiment involves a special-purpose computer called GF-11 that fills an entire 500-sq.-ft. room. Another computer, called RP-3, will consist of eight 8-ft. cubes arranged like a giant merry-go-round in a 35-ft. ring. But even these machines will be dwarfed by IBM’s most ambitious supercomputer, the TF-1, a behemoth whose specifications include 4,000 miles of internal wiring, 33,000 high-speed processing units and a single switching device measuring 80 ft. in diameter. When completed, the TF-1 should be capable of top speeds 2,000 times as fast as today’s supercomputers.

IBM’s real concern in the supercomputer market may be not Cray Research but Hitachi, Fujitsu and NEC. With their first generation of supercomputers, the Japanese made clear their intention to wipe out America’s 25-year lead. Today their fastest machines compare favorably with any supercomputer made in the U.S. In some applications they outperform the most advanced U.S. models. During a test comparing the newest single-processor Hitachi S-820/80 and a two-processor Cray X-MP, the Hitachi machine beat the Cray by about 10 to 1. Says Yukihiko Karaki, a professor at Senshu University in Tokyo: “Looking at these figures, one might say that Japanese users can do without Cray supercomputers.”

To date the Japanese have concentrated on speeding up the performance of their fastest processing chips. As a result, they now make the world’s most powerful sin- gle-processor supercomputers. But they have not, so far, begun linking large numbers of individual processors together. It is there, in parallel processing, that the U.S. still has the edge over the Japanese. A handful of small American manufacturers, including Bolt Beranek and Newman, NCUBE and Ametek Computer Research, have already started marketing parallel machines that can zip through equations at such blistering speeds that they threaten to put conventional supercomputers on the endangered list.

The sticking point with parallelism, however, is the software. Tens of thousands of man-years have been put into writing programs for traditional supercomputers. “Going parallel means starting over,” says Thomas Nash at the Fermi National Accelerator Laboratory. That is why the news from Sandia last week was so important. It confirmed that there are dramatic increases in speed to be achieved by breaking large problems into small pieces and solving them simultaneously. Says David Kuck, Chen’s former professor at the University of Illinois: “What’s going to happen in the next decade is that we’ll figure out how to make parallelism work.”

Making parallelism work will benefit not just supercomputer users but also those researchers in computer science’s other grand project, artificial intelligence. In fact, one of the most advanced parallel machines, a 65,536-processor computer called the Connection Machine, was built by researchers trained at M.I.T.’s Artificial Intelligence Laboratory. W. Daniel Hillis, the 31-year-old engineer who designed the computer, sees in it the first concrete evidence of what he views as an inevitable convergence of the two fields. “Supercomputing is an enabling technology for artificial intelligence,” says Hillis. “Just as you couldn’t build an airplane without first developing engines powerful enough to drive them, you can’t build artificial intelligence without faster supercomputers.”

Far more is at stake than the sale of a few multimillion-dollar machines. The country that leads the world in supercomputers and artificial intelligence will hold the keys to economic and technological development in the 1990s and beyond. Breakthroughs are waiting to be made in fields that range from genetic engineering to particle physics, from automated manufacturing to space exploration. There is even a chance that scientists will use the new computers to understand better the most complex machine of all, the human mind.

By Philip Elmer-De Witt. Reported by Thomas McCarroll/New York, J. Madeleine Nash/ Minneapolis and Charles Pelton/San Francisco

——————————
Just Dig While You Work
Where does the world’s foremost designer of high-speed computers get his inspiration? Apparently deep in a dirt tunnel beneath his Wisconsin home, according to John Rollwagen, the chairman of Cray Research. As Rollwagen tells it, Seymour Cray, the company’s elusive founder, has been dividing his time between building the next generation of supercomputers and digging an underground tunnel that starts below his Chippewa Falls house and heads toward the nearby woods. “He’s been working at it for some time now,” says Rollwagen, who reports that the tunnel is 8 ft. high, 4 ft. wide and lined with 4-by-4 cedar boards. When a tree fell through the top of the tunnel several years ago, Cray used the opening to install a periscope-equipped lookout.

For Cray, the excavation project is more than a simple diversion. “I work when I’m at home,” he recently told a visiting scientist. “I work for three hours, and then I get stumped, and I’m not making progress. So I quit, and I go and work in the tunnel. It takes me an hour or so to dig four inches and put in the 4-by-4s. Now, as you can see, I’m up in the Wisconsin woods, and there are elves in the woods. So when they see me leave, they come into my office and solve all the problems T’m having. Then I go back up and work some more.”

Rollwagen knows that Cray is only half kidding and that some of the designer’s greatest inspirations come when he is digging. Says the chairman: “The real work happens when Seymour is in the tunnel.”

COMMODORE VIC-20
“THE WONDER COMPUTER OF THE 1980s. UNDER $300.”

—WILLIAM SHATNER

“The best computer value in the world today. The only computer you’ll need for years to come.”

Read the chart and see why COMPUTE! Magazine1 calls the VIC-20 computer “an astounding machine for the price.” Why BYTE raves: “…the VIC-20 computer unit is unexcelled as a low-cost consumer computer.” Why Popular Mechanics says “… for the price of around $300, it’s the only game in town that is more than just a game.” And why ON COMPUTING INC. exclaims: “What is inside is an electronic marvel… if it sounds as if I’m in love with my new possession, I am.”

The wonder computer of the 1980s. The VIC-20 from Commodore, world’s leading manufacturer of a full range of desktop computers. See the VIC-20 at your local Commodore dealer and selected stores.

VIC-20 commodore COMPUTER Commodore Computer Systems 681 Moore Rd. King of Prussia, PA 19406
Canadian Residents: Commodore Computer Systems 3370 Pharmacy Ave., Agincourt, Ont., Canada, M1W 2K4

3-D trip inside a drawing, via computer graphics

Slip this display device on your head and you see a computer-generated 3-D image of a room before your eyes. Move your head and your perspective changes, just as though you were actually inside the room. Architects could use the device to draw buildings in three dimensions; realtors could use it to show buyers the interiors of homes without even leaving the office. Dr. Ivan Sutherland, University of Utah, invented the device, essentially a computer-graphics version of the old stereoscope.

view additional pages

The Second West Coast Computer Faire

By Chris Morgan, Editor

San Jose was the place to be last March 3, 4 and 5 for the Second West Coast Computer Faire. The Convention Center was easily able to handle the crowd of 14,169 who came to see the latest developments in personal computing.

A quick examination of some of the hundreds of manufacturers’ booths revealed some trends: floppy disks are on the increase, with new models being shown or promised by Heathkit, Apple, Radio Shack and many others; more and more personal computers are now being offered with built-in floppy disks; peripherals and add-ons are now available for a wide variety of computer buses.

I enjoyed the many special features of the show, particularly the excellent computer generated art on display in the lobby. The microcomputer chess tournament proved to be one of the hits of the show. Larry Wagner from Atari presided over the 3 day battle of the processors, taking time out to give me a guided tour of the tournament. The level of play was impressive, and the winning program, called SARGON, was a 16 K byte Z-80 assembler program written by a husband and wife team, Kathe and Dan Spracklen. It beat some highly touted com- petition. (A copy of the SARGON program is available for $15 postpaid from the Spracklens, 10832 Macouba PI, San Diego CA 92124.) I was impressed with the professional appearance of the show, which held its own with many of the established engineering and computing shows. The Third West Coast Computer Faire will be held this coming November 3, 4 and 5 in Los Angeles. Plan to see it if you can.”

Photo 1; Some of the 14,000+ crowd amble by a young hacker programming music on a Video Brain computer.

Photo 2: Robot trials at the Dynabyte booth, a popular attraction at the Second West Coast Computer Fa ire.

Photo 3: IBM’s booth, an auspicious addition to the show.

Photo 4: Ira Baxter’s chess playing system display, which competed in the Microcomputer Chess Tournament at the Faire.

Photo 5: Apple Computer’s new minifloppy drive.

Photo 6: Objective Design’s Larry Weinstein displays Star Wars graphics.

Photo 7: Heath’s new H27 dual floppy drive, scheduled to be available later this year.

Photo 8: Students from Mills College Center for Contemporary Music in Oakland demonstrate- a digital and analog hybrid music synthesizer system, one of many special exhibits at the Faire.

Photo 9: Cromemco color video unit displays chess program at the Computer Room of San Jose booth.

REMEMBER: ELEPHANT MEMORY SYSTEMS “NEVER FORGETS.”

MORE THAN JUST ANOTHER PRETTY FACE.

Says who? Says ANSI.

Specifically, subcommittee X3B8 of the American National Standards Institute (ANSI) says so. The fact is all Elephant™ floppies meet or exceed the specs required to meet or exceed all their standards.

But just who is “subcommittee X3B8″ to issue such pronouncements?

They’re a group of people representing a large, well-balanced cross section of disciplines—from academia, government agencies, and the computer industry. People from places like IBM, Hewlett-Packard, 3M, Lawrence Livermore Labs, The U.S. Department of Defense, Honeywell and The Association of Computer Programmers and Analysts. In short, it’s a bunch of high-caliber nitpickers whose mission, it seems, in order to make better disks for consumers, is also to make life miserable for everyone in the disk-making business.

How? By gathering together periodically (often, one suspects, under the full moon) to concoct more and more rules to increase the quality of flexible disks. Their most recent rule book runs over 20 single spaced pages—listing, and insisting upon—hundreds upon hundreds of standards a disk must meet in order to be blessed by ANSI. (And thereby be taken seriously by people who take disks seriously.) In fact, if you’d like a copy of this formidable document, for free, just let us know and we’ll send you one. Because once you know what it takes to make an Elephant for ANSI …

We think you’ll want us to make some Elephants for you.

ELEPHANT HEAVY DUTY DISKS.

Distributed Exclusively by Leading Edge Products, Inc., 225 Turnpike Street, Canton, Massachusetts 02021 Call: toll-free 1-800-343-6833; or in Massachusetts call collect (617) 828-8150. Telex 951-624.