Articles 1

20 Most Important Chips (Article from 1995)

All the chips on this list, obscure as some are, had a significant influence on the evolution of personal computing. So what does it take to make a computer today? Mostly, it seems, acronyms: a CPU, some RAM, a handful of EPROMs, a DSP, and a PCI bus.

Intel 1103

In 1970, Intel created the 1103--the first generally available DRAM chip. By 1972, it was the best-selling semiconductor memory chip in the world. Today, you would need more than 65,000 of them to put 8 MB of memory into a PC.

Intel 1702

In another brilliant stroke of naming, Intel created this, the first EPROM, in 1971. When you say "firmware," smile and think of the 1702.

Intel 4004

In 1971, Busicom, a Japanese company, wanted a chip for a new calculator. With incredible o verkill, Intel built the world's first general-purpose microprocessor. Then it bought back the rights for $60,000.

The 4-bit 4004 ran at 108 kHz and contained 2300 transistors. Its speed is estimated at 0.06 MIPS. By comparison, Intel's latest microprocessor, the P6 , runs at 133 MHz, contains 5.5 million transistors, and executes 300 MIPS.

Intel 8080

If you drive, your life probably depends on this chip. Introduced in April 1974, the 8080 was first widely used as a traffic-light controller. It found its way a year later into the world's first personal computer: the MITS Altair.

MOS Technology 6502

What do a Nintendo set and a BMW have in common? The 6502. At $25 (compared with $375 for a comparable Motorola part), the 6502 was such a steal that a talented but cash-poor whiz kid from Silicon Valley, Steve Wozniak, chose it for his new personal computer, the Apple I.

Zilog Z80

Remember Tandy's TRS-80 Model I? Remember CP/M? They were both built on the Z80.

Intel 8086 and 8088

Enter the King. In June 1978, the 8086 debuted. Today it stands for the world's most popular microprocessor standard: the x86 architecture. A year later, Intel introduced a slight variation, the 8088, that could use 8-bit components, enabling the manufacture of inexpensive systems. For that reason, IBM chose the 8088 over the 8086 for the original IBM PC, even though the 8088 was slower.

Intel 386DX

The 386 heralded the beginning of a new age--the age of multitasking. Introduced in October 1985, the 386 was the first "modern" x86 processor that was capable of running today's multitasking OSes, GUIs, and 32-bit software.

The 386 introduced an enhanced microarchitecture while maintaining full backward compatibility with earlier x86 processors. This was accomplished with two memory-addressing modes: real mode, which mi rrored the way memory is addressed by the older x86s, and a new protected mode that took full advantage of the 386's 32-bit enhancements.

Intel Pentium

The Pentium swept through the PC industry faster than any of Intel's previous chips. Although Intel's 486DX (April 1989) integrated an FPU and was much faster than the 386, it was the Pentium that introduced the next leap forward in the x86 microarchitecture: superscalar pipelines. Skeptics said a CISC architecture couldn't do it. The Pentium proved otherwise.

AMD 386DX

Let the price wars begin. When Intel's original 16-MHz 386 was introduced in 1985, it cost $299; more than five years later, it was still commanding the relatively high price of $171, and the 33-MHz version fetched $214. AMD's 386DX/40 appeared in March 1991 at $281, but within a year its price plunged 50 percent to $140. Street prices of PCs, which follow chip prices, fell by as much as $1000. The market for Windows-capable PCs expand ed by 33 percent.

Motorola 68000

More than any other, this is the microprocessor that helped establish the GUI. In 1983, four years after its introduction, it appeared in Apple's Lisa, a unique computer but a commercial flop that nevertheless paved the way for the Macintosh in 1984.

Mips R2000

The R2000, introduced in 1986, was a 32-bit CPU with 110,000 transistors. It powered the first generation of RISC workstations and servers. The original version, clocked at 8 MHz, executed about 5 MIPS and had a separate FPU.

Sun Microsystems SPARC

In July 1987, Sun announced an open RISC architecture. The idea was to encourage multiple sourcing and lively competition that would spur performance and spread the SPARC standard far and wide. Eight years later, SPARC workstations and servers dominate their markets.

IBM/Motorola PowerPC 601

Although few doubted the power of the Po werPC architecture, many thought the politics of the IBM/Motorola/Apple relationship was going to be unmanageable.

In less than two years, it has spawned the world's most popular RISC platform: the Power Macintosh.

Chips & Technologies AT Chip Set

IBM is not known for its approach to open systems. So, while it was actively resisting the cloning of its PC architecture, C&T was introducing its AT Chip Set. With only five chips, C&T duplicated the core logic of about 100 chips in IBM's system. All a clone maker had to do was add a 286, a Phoenix BIOS ROM, and some memory to create a PC. Take that, Big Blue.

Amiga Agnes/Denise/Paula

It's not a rock group: This was the advanced chip set that powered the world's first multimedia computer: the Commodore Amiga 1000. In 1985, these three chips could do tricks that today's PCs and Macs still can't do--such as display multiple screens with independent pixel resolutions and bit depths on a single monitor.

Commodore SID

You can get remarkable results when you tell an engineer to do what he thinks is right. Take SID (Sound Interface Device), for example. In 1981, Bob Yannes was told to design a low-cost sound chip for the upcoming Commodore 64. He would end up creating an analog synthesizer chip that redefined the concept of sound in personal computers.

Yamaha OPL-2

Tweet. Beep, beep. Name that tune! The original IBM PC's sound capabilities were practically nonexistent--a simple beeper that could produce a limited range of square-wave tones. Yamaha's OPL-2 enabled vendors such as Ad Lib and Creative Labs to introduce plug-in sound boards with reasonable (but not great) sound. Today, nearly all PCs come with a sound board.

S3 911

Because PCs originally had character-oriented displays, screen performance drastically bogged down when running Microsoft Windows and graphical applications.

IBM's 8514 chip and its spin-off s provided some improvement, but the market broke wide open in 1991 when S3 introduced the 911, which integrated GUI acceleration and VGA compatibility on a single chip.

Intel Mercury

The PCI (Peripheral Component Interconnect) bus is the most important enhancement to the PC architecture since the ISA bus, and Mercury was the first implementation. Today even Apple has adopted PCI to replace the NuBus.

Top of Page

Glossary of Microprocessor Architecture Terms

Bus Unit:
The Bus Unit is the place where instructions flow in and out of the microprocessor from the computer's main memory.

Instruction Cache:
The Instruction Cache is a warehouse of instructions right on the chip, so that the microprocessor doesn't have to stop and look in the computer's main memory for instructions. This quick access makes processing fast as instructions are 'fetched' to the Prefetch Unit where they are put in the proper order for processing.

Prefetch Unit:
The Prefetch Unit decides when to order data and instructions from the Instruction Cache or the computer's main memory based on commands or the task at hand. When the instructions come in the most important task for the Prefetch Unit is to be sure all the instructions are lined up correctly to send off to the Decode Unit.

Decode Unit:
The Decode Unit does just that - it decodes or translates complex machine language instructions into a simple format understood by the Arithmetic Logic Unit (ALU) and the Registers. This makes processing more efficient.

Control Unit:
The Control Unit is one of the most important parts of the microprocessor because it is in charge of the entire process. Based on instructions from the Decode Unit, it creates control signals that tell the Arithmetic Logic Unit (ALU) and the Registers how to operate, what to operate on, and what to do with the result. The Control Unit makes sure everything happens in the right place at the right time.

Arithmetic Logic Unit (ALU):
The ALU is the last stage of processing in the chip. The ALU is the smart part of the chip that performs commands like adding, subtracting, multiplying and dividing. It also knows how to read logic commands like OR, AND, or NOT. Messages from the Control Unit tell the ALU what it should do and then it takes data from its close companion, the Registers, to complete the task. This is really where 2 finally gets added to 3 in our example.

Registers:
The Registers are a mini-storage area for data used by the Arithmetic Logic Unit (ALU) to complete the tasks the Control Unit has requested. The data can come from the data cache, main memory or the control unit and are all stored at special locations within the Registers. This makes retrieval for the ALU quick and efficient.

Data Cache:
The Data Cache works very closely with the "processing partners," the ALU & Registers, and the Decode Unit. This is where specially labeled data from the Decode Unit are stored for later use by the ALU and where final results are prepared for distribution to different parts of the computer.

Main Memory:
This is the big store house of data located within the main computer outside of the microprocessor. At times the Main Memory may send in data or instructions for the Prefetch Unit, which often get stored at an address in the Instruction Cache to be used later.

WHAT EVER HAPPENED TO... THE TI 9900 CPU?

by Stan Veit
Originally published in the September 1996 issue

Now that super chips are being produced by companies other than Intel
and Motorola, I am reminded that the success of new processors can
rely on more than technological innovation. The silicon graveyards are
filled with technically superior CPUs that lost out because of poor
marketing or because the manufacturer misread the buying habits of the
public.

In the early days of the PC industry, the Intel 8080 family received a
huge boost because of the public's early adoption of the Altair/S-100
series; Zilog's Z-80, which was compatible with the 8080 CPU,
continued that success. Similarly, Motorola's 6800 chips were used in
the widely popular South West Technical Products PC and many others.

A Different View

Texas Instruments (TI), the world's largest semiconductor
manufacturer, developed the TMS9900, a true 16-bit CPU that was quite
advanced for its time, having capabilities that the Intel and Motorola
8-bit CPUs lacked. Unlike TI, chip makers Intel, Zilog, and Motorola
were not eager to become computer manufacturers. They made development
systems, but these were not designed or priced for public use. Most
chip manufacturers did all they could to help computer makers improve
the systems that utilized their chips, but TI gave little help to
outsiders.

TI did develop the TI 990/4 and 990/5 minicomputers, which used the
TMS9900 processor, but they were too expensive to attract developers
who would write software for the processor. Although TI did little to
encourage second sources for the TMS9900, TI itself was a second
source for Intel, producing the TMS8080A, a version of the Intel 8080.
In fact, much of TI's PC competition was powered by chips it had
manufactured.

As a corporation, TI had a different view of the PC market. It
envisioned a product that would be used in the home, at school, and on
the job, just as its calculators were. This machine would be filled
with TI chips, run TI software, and earn millions for the company.
Since such a machine did not yet exist, TI had to design one--the
Texas Instruments 99/4.

In the spring of 1979, I went to Boston to see the 99/4--I was hoping
to sell it at my Computer Mart store in New York. I also had thoughts
of becoming a software developer for graphics-based games. Priced at
$1,200, the 99/4 had highly polished metal parts, 16K of RAM, TI BASIC
in ROM, a 13-inch color monitor, and a thin keyboard. The 40 keys on
this prototype were of a style that came to be called "Chiclets," and
the monitor displayed 24 lines of 32 characters. The system also had a
built-in slot for plug-in, solid-state software modules, which did not
yet exist. The plug-in program modules were to hold extra RAM to run
the software.

Interestingly, the TI engineers had crippled its 16-bit CPU by running
it in a 8-bit bus. This permitted them to use fewer memory chips and
reduce costs. Later, IBM followed the same path with the IBM PC, using
the 8088 microprocessor rather than Intel's 16-bit 8086.

Back to the Drawing Board

I liked the TI 99/4, but thought the price was a little steep. So did
the rest of the world. Shortly after it was released, TI recalled the
99/4 and went back to the drawing board to bring down the costs. The
new TI 99/4A required fewer chips, due to large-scale integration, and
it came with a real keyboard. TI priced it without the monitor and
provided plug-ins for expansion. By the time of the 99/4A's release,
some software for it had appeared, and TI encouraged third-party
software developers to write for the new system. However, the
royalties TI demanded were discouraging; ultimately software
developers lost interest in the 99/4A.

In late 1982, TI re-engineered the motherboard to put most of the
"glue" chips into a single package. The new machine was called the QI,
for "Quality Improved." With a new and improved system, TI had a
system that could compete with the Apple, Atari, and Commodore systems
of the day. It also had a better product for a small but growing group
of devoted users. The 99/4 series users were among the most devoted
fans you could imagine. But the company did little to support users,
and except for Computer Shopper and a small magazine called 99er, the
computer press ignored the systems.

When all other manufacturers turned to floppy disks to expand their
machines' usefulness, TI came out with the expensive Expansion Box as
a way to add a floppy drive to the 99/4. It was designed like a piece
of equipment meant to meet military specifications; the box was made
of a thick aluminum plate, a heavy-duty design that was obviously
unnecessary for home computers. Compare this with the Apple II, which
needed only a small plug-in interface board to connect a floppy disk.
Once again, TI had to provide a fix to the system because it did not
understand the market and had to learn the hard way.

In January 1982, TI was poised to throw all its marketing efforts
behind the TI 994/A. At the time, the success of TI's home computer
meant more to the company than just another item in its huge line of
electronic equipment. It was counting on the TI 99/4A, and its
successors to be the major consumer of the company's own chips. This
was TI's core business, and the strategy would have succeeded if TI
hadn't run into a stone wall--Jack Tramiel, the president of
Commodore.

The troubles started with the introduction of the $300 Commodore
VIC-20. Although the VIC's capabilities were below those of the TI
99/4A, its introduction cut into the TI system's market share.
Commodore followed up the VIC-20 with the introduction of the
excellent Commodore C-64 and started selling them through discount
stores like Kmart. The cost of the C-64 quickly dropped from the $595
introductory price to about $400.

Rebate Wars

In August 1982, TI issued a $100 rebate on the 99/4A and a price war
was on. Atari joined the battle with rebates on its 800-series
systems, but those systems' list prices were way above the
free-falling cost of the Commodore and TI systems. By February 1983,
TI again cut the price to dealers, and the cost of a TI 99/4A dropped
to about $150. At this point, the Commodore 64 cost about $350, and
the VIC20 cost less than $100, including peripherals.

TI suffered a massive blow when the power supplies in the 99/4A proved
to be defective. Retailers had to stop selling the systems, and TI had
to replace thousands of power supplies. Sales were dead, and Commodore
announced a $50 trade-in rebate for any computer, even nonworking
ones. This dropped the street price of a Commodore 64 to less than
$300.

Commodore later announced additional cuts in C-64 dealer prices,
bringing costs down to $200 and causing the retail price to fall to
about $250. Commodore also cut software prices by 50 percent and
lowered peripheral prices, too. Tramiel slashed Commodore's production
costs so drastically that in the end, the C-64 cost about $100 to
manufacture.

Meanwhile, TI was bleeding dollars with every price cut. In June 1983,
after counting second-quarter losses of $100 million, Bill Turner,
president of TI's consumer division, announced that Texas Instruments
was bowing out of the home-computer market.

The day of the announcement was called Black Friday by the loyal TI
99ers, but the announcement lowered prices on the TI 99/4A to less
than $100 in stores all over the United States. Thousands of people
who had considered buying a home computer suddenly rushed out to buy a
TI 99/4A at its closeout price. Although many of these systems ended
up as doorstops, they provided the first look at personal computing
for many users.

The death of the TI 99/4A did not quite kill the TMS9900 family of
CPUs, however. Some minicomputer manufacturers continued to use it,
but the handwriting was on the wall. Soon other companies were making
faster 16- and 32-bit chips. But TI 99/4A fans carried on, and some
small companies continued to provide software and peripheral support
for years afterward.

In hindsight, it's easy to see that Texas Instruments learned little
from the disastrous TI 99/4A-Commodore 64 conflict. Soon after, the
company released the TI Professional Computer, a rival to the IBM
PC.This non-IBM-compatible 8088 machine, which used special versions
of MS-DOS, CP/M-86, and application software, had a good keyboard and
fantastic color graphics. But its proprietary architecture killed it.
Since then, TI has focused on its calculator, chip, and printer
businesses. Interestingly, it was the people who left Texas
Instruments to found Compaq Computer who took on IBM and won.

PC Pioneer Gary Kildall

Kildall lost to Gates with IBM Personal computer giant remembered for deal he didn't make

Personal computer pioneer Gary Kildall, who but for a single failed business deal might have enjoyed the wealth and fame of Bill Gates, died in 1994 aged 52 at the Community Hospital of the Monterey Peninsula

Kildall apparently was taken to the hospital after suffering a concussion in a fall, said Thomas Rolander, a longtime friend and former business associate of Kildall. Rolander said evidence indicates Kildall suffered a fatal heart attack. It is unclear if the two conditions were related.

Kildall's career spans the history of the personal computer, which he was instrumental in popularizing in the 1970s.

"Gary's technical contributions in the beginning days of microcomputing were order-of-magnitude enhancements to the capabilities with which we were working," said Jim Warren, a Woodside consultant who played a key role in early microcomputing. "The were enhancements both in technical power and in equitable consumer-oriented pricing and support practices."

In 1972, Kildall was an associate professor of computer science at the U.S. Naval Postgraduate School in Monterey and a consultant for Intel Corp., which the year before had created the world's first microprocessor chip. Kildall wrote a version of the PL/I programming language that worked on the chip, the Intel 4004. A year later, frustrated with the difficulty of making the 4004 work with disk drives and other computer components, he wrote the first personal computer operating system.

A PC Breakthrough

The program, called Control Program for Micro-Computers and shortened to CP/M, offered hobbyists a way to use their microcomputers, as they were then called, in the same way as larger minicomputers and mainframes. Before, the computers were programmed in laborious ways, like flipping switches on the front panel of the machines. With CP/M, they could type instructions on a keyboard, store data on a floppy disk or tape recorder and view results on a screen or printer.

Digital Research, the company started in 1976 by Kildasll and his first wife, Dorth McEwen, sold CP/M for $75 each. Kildall, who disliked business, said in a 1981 interview that he hoped "just to support my computer habits" with the proceeds.

But the typical minicomputer operating system at the time sold for at least $10,000, and Intel's own operating system for microcomputers cost $800. CP/M soon became the standard operating system for personal computers, which could be bought for as little as a thousand dollars. By 1981, Kildall was one of the best known figures in the $2 billion personal computer business, and his $10 million company had sold 250,000 copies of CP/M.

Negotiated with IBM

However, Kildall is probably best remembered for being on the losing end of one of the biggest deals in computer history.

In 1980, IBM contacted Digital Research, hoping to persuatde it to produce a new version of CP/M for the personal computer IBM was secretly developing. Kildall didn't think much of IBM"s chances but met with the company anyway.

"IBM wanted to take the market away from Apple, and they looked at them and saw that the SoftCard (a CP/M add-in card for the Apple II) was an important part of it," Kildall said in a 1991 interview.

Negotiations went badly, Rolander said. IBM wanted Digital Research to sign a non-disclosure agreement but refused to sign one in return. IBM wanted to pay a flat fee for CP/M, with no royalties, and change the software's name.

Silicon Valley legend has it that Kildall, a passionate private pilot, missed a crucial meeting because he decided to go flying instead. While Kildall did fly that morning, Rolander said, he attended the afternoon meeting.

IBM decided to hedge its bets. During a visit to tiny Microsoft Corp., to obtain a version of its BASIC programming language, IBM inquired if the company also could provide an operating system.

Microsoft moves in

Even though he didn't have one, Microsoft founder Bill Gates readily agreed to IBM's request. He bought a CP/M clone called DOS from Seattle Computer Products, a company run by a friend of Gates, for $250,000. That program became MS-DOS, proably the most widely used software in the world, and helped turn Gates into a billionaire.

Kildall had earlier sued Seattle Computer Products for copyright infringement. When he confronted IBM with the fact, IBM responded that it would agree to license CP/M as well -- if Kildall agreed never to sue. He did, only to discover when the IBM PC was introduced that the price of DOS was $40, while the price of CP/M-86 was $200 more.

"It was only through inadequately sharp business hustling that MS-DOS took the IBM cake when, by rights, CP/M should have done so," Warren said.

But hard-nosed business was not Kildall's style.

"Basicly I am a gadget-oriented person," Kildall said in 1981. "I like to work with gadgets, dials and knobs. I'm not a very competitive person. I'm forced into it."

Kildasll remained active in the industry until his death. He was Digital Research chairman until 1991, when Novell Inc. bought the company. He started an early multimedia company in Monterey in 1985, and later moved to Austin, Texas, to persue the field. He recently returned to Monterey and spent the last year and a half writing an unpublished book on the computer industry called "Computer Connections."

Kildall was born in Seattle on May 19, 1942, and studied computer science at the University of Washington, eventually earning a Ph.D. He then took his post at the Naval Postgraduate School.

Kildall met McEwen while in high school. The two married in 1963 and were divorced 20 years later. Kildall married his second wife, Karen, in 1986. They were recently divorced.

Kildall is survived by two children; Scott, of San Fransisco, and Kristin, of Seattle; his mother, Emma; and a sister, Patti Guberlet, both of Seattle.

Deep Blue's computer chess victory creates deep puzzles about humanity

Ed Roberts Interview with Historically Brewed magazine

Ed Roberts is the Father of the Personal Computer. Now, you can argue the point, but it is generally accepted that the MITS Altair, circa 1975, was the first mass produced, commercially successful personal computer, and Ed Roberts, with some help, masterminded its creation and success. Here is a short version of that great American success story, as it was printed in our very first issue -

The Altair Leslie (Les) Solomon was the Technical Editor for Popular Electronics during the summer of '74 and he was looking for a good computer article and project to print. Both he and Editorial Director, Arthur Salsberg wanted to publish a piece on building a computer at home. Solomon had received some articles, but they were not what he was looking for. "A rat's nest of wires," as he would describe them. But, Solomon encouraged his writers to send in their best ideas. Ed Roberts was one of "Uncle Sol's" writing contributors. A man who loved to fool with gadgets and electronics, Roberts started a small electronics company in Albuquerque, New Mexico in 1968. MITS (Micro Instrumentation Telemetry Systems) mostly sold radio transmitters for model airplanes through the mail. But, by the early 70's, MITS was selling calculator kits and doing fairly well. At the end of 1973, the calculator market changed drastically, other companies were selling fully assembled calculators for below $50, while Roberts' kits were $99.95. He had to think of something quick or go broke. He had toyed with the idea of developing a computer kit before, but never followed up on it. Now, he decided to go for broke. If this didn't work, then he would just close up shop. Roberts decided on the Intel 8080 chip for his project, rejecting the older 8008 and new Motorola 6800. He was able to get an excellent deal on the chip in volume - $75 a piece for a $360 chip! By mid-1974, Solomon had decided on supporting Roberts' article and kit. He staked the reputation of PE on the expertise of MITS. In July 1974, Radio Electronics had published an article on a 8008 based computer kit called the "Mark-8". Les Solomon needed an 8080 based project to beat out RE. MITS worked feverishly on the computer, creating an expandable main circuit board that had a data bus with 100 separate paths. It was capable (in a miniature way) to do anything that a large mainframe computer could do. Les Solomon's daughter, Lauren, gave it it's name "Altair" because that was where the Enterprise on STAR TREK was going that night. He had asked her for a name idea and had asked what they called the computer on STAR TREK. "Computer," she said. Altair was the better name. The Altair kit appeared on the cover of the January 1975 issue of Popular Electronics. In kit form, MITS offered the basic model with 256 bytes of RAM, standard binary switches and LEDs on the front panel and power supply for $400. Hoping to sell around 200 kits, Ed Roberts and MITS were overwhelmed to receive thousands of pre-payed orders. Electronic hobbiests were willing to have paid the $360 just for the chip itself, so why not get an entire computer for $400? It took MITS almost a year to catch up the orders. Over 10,000 Altairs were sold by MITS. The Altair was the first commercially successful computer ever. It started the personal computer revolution which has since consumed our planet. Imagine the world just 18 years ago (almost 21 years ago now, DG) when there wasn't a computer in every pot.

- O.K., well that pretty much sets you up for our interview if you weren't already familiar with the story that began it all. Ed Roberts is now a doctor and lives in a small Georgia town named Cochran. I drove to his office one weekend earlier this year to talk personally with a man who I have read so much about. Our informal discussion spread well over two and a half hours, here are the highlights:

"HB": Do you remember the "T.V. Typewriter" and did it influence you at all? Were you excited about its appearance?

Ed: I remember the T.V. Typewriter and we were building something like it if I'm thinking about the same thing - we built a computer terminal which we actually sold off and on up to when Pertec took over. We used a Burroughs display which was a neon display that had dot matrix output. It had a modem built into it and you could put the phone receiver right into the top. It was something similar to what Lancaster had written about. He was a real pioneer with mail order kits and a lot of the stuff that he did served as a model for M.I.T.S. The one thing that I remember he did that really impressed me was sometime in the mid '60's. He created a decimal counter made out of small scale integrated circuits that could count from 0 to 10. You could stack these things to make a bigger counter. It was really neat even though it doesn't sound like much these days. We used a Teletype more than anything else on those initial machines. Those old ASR-33's with the paper tape gave you a way to input and output as well as a way to load data. You could buy them used for only a few hundred dollars at the time.

"HB": That was kind of a gift from heaven at that time - the ultimate peripheral!

Ed: Yeah, you got your line printer, your input / output device and your mass storage.

"HB": When you first decided that you wanted to create a personal computer, what were some of the things that you thought it must have, and did you ever say I definitely don't want it to be like this?

Ed: The basic ground rules for a personal computer from a technical standpoint is that it had to be a real, fully operational computer that was fully expandable and at least in principal could do anything that a general purpose minicomputer of the time could do. "Minicomputer" was the term them and referred to any 16-bit or 8-bit machine. And, those were the ground rules. We wanted to make a machine that was, from a users stand-point, not degenerative at all. The main difference between our machine and where others were, is that we used microprocessors and everything was the latest state of the art. We never used core memory even though we did look at core. At the time we began work on the Altair, core memory was still significantly cheaper than IC based memory.

"HB": You used static memory first and then dynamic, right?

Ed: The original machines had 256 bytes of static, and that was the minimal configuration. That board in the machine came that way and was expandable up to 2 Kb. We later developed a 2 Kb, 4 Kb and ultimately a 16 Kb memory board before I left; there may have even been a 32 Kb board. We did sell both static and later dynamic memory.

"HB": HCS has a few MITS Altair memory boards. OK, did any particular type of minicomputer influence you? I've read that it was a PDP that you most favored.

Ed: We had a Nova 2 by Data General in the office that we sold time share on, and as a matter of fact that was how we got into building a little terminal. The front panel on an Altair essentially models every switch that was on the Nova 2. We had that machine to look at. The switches are pretty much standard of any front panel machine. It would have taken forever if we would have had to re-decide where every switch had to go. These guys had already figured it out. That's pretty much the way technology works. If you were going to build a machine today that would allow front panel control, it would look just like an Altair, or Nova, PDP-10 or any of those front panel computers. There are certain switches you need to get an address in, to get data in, to single step it, look at output and reset it.

"HB": What were the series of products that MITS had? You had your radio remote control devices, then went to calculators and then computers right?

Ed: The very first product that we made and was featured in Popular Electronics, was an optical communicator; it was a laser communicator, but it used infrared diodes and wasn't really a laser. We had lenses that we produced ourselves, and it was sold as a kit. They had a range of 500 or 1000 feet. It was actually a pretty good little product. We sold 100's of them which was pretty good for only two of us working out of a garage. There was only one calculator ever featured in Popular Electronics I believe; we made the 816 a & b, the 1440, which had square root, we made a machine which had fourteen digits and square root and memory, also we made a hand held scientific machine. I think we made two or three different models of those. We also produced a scientific metric converter, logic analyzer, function generators and a whole bunch of products over the years.

"HB": And as far as the Altairs, let's see, there was the original 8800, 8800a, 8800b, 680, 680b and your turn-key 8800b and 680b. What other peripherals?

Ed: When I left MITS in 1977, we offered about forty different computer products. We offered two or three memory variants, I/O boards, serial and parallel cards, disk controllers and D to A converters. I think we also had a controller for the really big drives that we produced for somebody. We had another machine and I'm not sure what Pertec did with it after I left, but it was a self-contained computer that came with either a 40 digit or 64 digit display. It was sort of an "Apple" before the Apple ][ came out. As a matter of fact, it was somewhat like what Processor Technology came out with later. With the Sol computer, they used the "Altair Bus" and incorporated a keyboard and the display control in one box. It was still an Altair bus compatible machine. That's a bit of a sore point that everyone changed the name of the bus.

"HB": Yeah, I've read a lot about that.

Ed: Nobody wanted to steal the name, they did that because they hated to give us credit every time they talked about their own product. A bunch of vendors got together and decided to call the bus "S-100". We should have copyrighted the name or patented the bus, but we never did that. Anyway, the Sol was an Altair based bus machine and really was an "Apple" before the Apple ][. It was a good machine and they were good competitors. As a matter of fact, the company that I think about out of all of the old companies more than any other is Processor Tech, for a couple of reasons. Number one, they were good competition, building good products, mostly add-on products. They were the first people to build add-on products for the Altair that I felt were of any quality. There was tons of junk stuff being built, but those guys did good work.

"HB": I've always heard good things about Processor Tech. We have a Sol, and it is an interesting twist on the Altair's design.

Ed: And the funny thing is, that I bet every time I have made that comment in any in-depth interview, the only story that is ever told about MITS and Processor Tech, is about the first ever Altair Convention in Albuquerque. It was David Bunnell's idea (MITS' Director of Marketing), and I said "Dave, this is crazy, we're not going to get people to come all the way to Albuquerque to go to this convention."

"HB": But they did!

Ed: They did! Tons of them! He was absolutely right. Anyway, Processor Tech showed up and they had a suite up on the ninth floor of this Airport Marina Hotel where the convention was, and they had a sign down in the lobby that said "Processor Technology". David Bunnell tore up their sign or made them take it down or something. I was unaware of this and didn't have anything to do with it. It created more flack to hurt us then if we had just ignored it.

"HB": Yes, I too have read that you were responsible for that.

Ed: David worked real hard on the convention and it made him mad that those guys were going to get some benefit from what he had done. But anyway, that's the only story that gets printed. Nothing that I have ever said good about Processor Tech ever gets printed. They identified some of the problems in the Altair and created good products to fix them. It's hard to really say anything bad about them. You know, it's the stuff that sells when you're writing an article . . .

"HB": All the bad stuff - the dirt! Yeah, I'm not trying to get at any of that. For me, it's just interesting to talk to you and to hear what you have to say about the events of the time. I'm trying to constantly think of somethi ng to ask you that is different from anything published before. The basic story has been told so many times already. I just remembered one somewhat silly question that I wanted to ask you - what type of computer do you use now?

Ed: All of the machines that I have now are PC based machines. At the house, I've got a 486DX4/100, NEC Versa portable 33 and a Z-Lite 486 black and white, which I really like.

"HB": I remember reading somewhere that you really liked your "gadgets" too! I guess that hasn't changed over the years!

Ed: That's true! Isn't it true about most people who like computers though? I remember I made a comment at that show which got some publicity and I haven't heard much about it recently. I don't know if you remember when Carter was running for President, but he made the comment that he "used to lust after women" or something like that. Well, all of us who were connected with computers then lusted after computers. Computer users now are very different kinds of people, mostly business users. To have a computer in the old days was better than sex; it was really something exciting.

"HB": I can tell you from my experience working at CompUSA, that the majority of people buying computers - don't even know why they're buying one, they just think they "should" have one! That's not really a good reason to buy one, but that's what is going on in 1995.

Ed: Yeah, the difference is that then, people lusted after the machines and wanted them. To a large extent, they didn't know what they were going to do with them, but they knew that they wanted one. These were mostly people who used computers professionally.

"HB": What do you still own? Did you keep some of your favorites?

Ed: There are two Altairs in here and both are "brand new". They have never been used, just checked for power. There is an "a" model and a "b". They were wrapped up until recently. We had a Japanese crew come in and take some pictures.

"HB": Do you have any favorite user stories you would like to share?

Ed: There was a dentist in Chicago who was one of our very first customers. He wanted to use the Altair to control a massive model railroad. And that was a real eye opener to us that people were coming up with applications and ideas that we had never even imagined. It makes me think of that old adage that if you gave and infinite number of monkeys an infinite number of typewriters, that one of them would write King Lear!

"HB": And with the computer - sooner or later, someone will come up with something amazing!

Ed: And that was something that always intrigued me. When I was going to school between '65 and '68 at Oklahoma State, which was a very "forward thinking" school believe it or not, they had an IBM 1620 in an open lab. No one monitored the machine. This was a big machine with a lot of money invested into it. It was open to engineering students and we would go down there and just put our name on a roster to use it. It was fantastic! And that had probably more impact on my feelings later on about computers than anything else. Computers had always been sort of "Mecas" up to that point.

"HB": Did you ever have one of the minicomputer companies buy an Altair just to check it out?

Ed: It must have been around late 1975 or early 1976, I remember it was before we moved into our new building, but anyway, we met with IBM in the MITS' executive lunch room. The lunch room was located behind my back office, through an alley and in the Dairy Queen! That's what we all called the MITS' executive lunch room! Anyway, IBM showed up with a bunch of lawyers and wanted us to be a witness in their case against Memorex I think it was. Apparently Memorex was suing IBM for monopolistic practices. IBM came up with some figures that showed MITS and our Altair to be increasing the supply for computers in the world by 1% each month. They wanted to use us in court to show that we were producing more computers than even they were at that time! I think that was a big reason why they came to MITS, but also I think they came to check out what we were doing.

"HB": Can you remember the most famous or unforgettable person who bought an Altair?

Ed: The guy who did some of the special effects for Star Wars, or I believe Star Wars, came out to MITS and bought some equipment. I can't remember his name. Lets see, we also sold stuff to the Secret Service, the FBI and the CIA. They were bought up by all kinds of people.

"HB": For the most part, they could take the place of any minicomputer at the time.

Ed: It took the industry a long time to realize that microprocessors and microcomputers were also very useful. People went through this thing where there were "supercomputers", there were "computers", there were "minicomputers" and then there were "microcomputers". Everyone assumed that when you said "microcomputer" you were talking about a performance thing. A microcomputer was really a technology and that was one of the arguments which went on at the time - the term "microcomputer" didn't same a thing about it's performance, they were missing the whole point. Microprocessors were approaching the power of minicomputers even with the Intel 8080.

"HB": I guess really the only thing that minicomputers had over micros at the time was the fact that some were 16 and 32 bit which made them somewhat more substantial. Right now with the high end Pentiums and PowerPC's, you essentially have the raw processing power of the supercomputers of fifteen years ago which cost millions of dollars.

Ed: I made a prediction actually, which everyone at the time thought was a joke, that the personal computer would destroy IBM. They made a big hit with the PC, but almost went belly up here recently.

"HB": I remember that, they laid off thousands all over the world, but somehow they seem to have made a come back. Their "ThinkPad" is extremely popular and their "Aptiva" systems are doing modestly well.

Ed: There's nothing "Gee Whiz" about their desk machines is there?

"HB": No, nothing that I have seen. In my personal opinion, I still think Apple is the biggest innovator out there. I'm no engineer, but they still seem to consistently come up with the new original ideas and everyone else copies them.

Ed: You know, Dave Bunnell, who still keeps up with most of the new technology says that the thing that is kind of interesting to him, is that if he goes to a meeting with PC users or a meeting with people who use Apples; the Apple users look a lot more like the early MIT's customers than the PC users. With the Apple users, you can see the same enthusiasm and excitement they have about their machines. They seem to "love" their computers which is a different kind of mentality.

"HB": I think you're right, Mac people are still very enthusiastic about their computers. I guess it has to do with being a minority or something.

Ed: Yeah, they're still kind of considered the "little guys" amongst PC users.

"HB": And maybe back in the Altair days the hobbyists and electronics guys were kind of a minority too - a specialty group of people who had to stick together because everyone else thought they were sort of weird.