The History of Personal Computers – Part 1: The East Coast Version

RSS

My version of the origin of personal computers.

Have you ever wondered where PCs and all their derivatives came from?  If not, I am going to tell you anyway, at least my version of it.  Of course, the source was the development of the microprocessor in the early 1970s. 

One of the things that really ticks me off is to hear that the PC had its start in Silicon Valley.  Naturally, much of the development occurred there but lots went on outside of the west coast.  If you read the book Fire in the Valley (Osborne/McGraw Hill, 1984) you get the impression that the whole PC and hobby computing movement occurred in Silicon Valley and that Steve Wozniak and Steve Jobs invented the PC and Bill Gates invented software.  But there is more to the story than that. 

My first remembrance of a hobby computer movement was in the late 1960s when I first subscribed to Stephen B. Grey’s Amateur Computer Society newsletter.  The microprocessor hadn’t been invented yet and the big computing boom was in minicomputers made by Digital Equipment Corporation (DEC), Data General and others.  They were made with ICs and discretes and were horrendously expensive.  DEC made the cheapest mini called a PDP-8E for about $9000.  A small hardy group of electronic hobbyists were trying to build their own computers out of cheap digital ICs (RTL and DTL mostly) and surplus magnetic core memories out of old IBM 1620 computers.  Solid state memory was not available yet. The ACS newsletter served as an information source on ICs, memories and potential I/O terminal devices.

In the late 1960s and early 1970s, a couple of computer kits came into being.  One was the Kenback made with 7400 series TTL.  An educational training computer by Fabri-Tek called the BiTran 6 was available to schools and the military.  I developed a computer kit for McGraw Hill’s National Radio Institute (NRI) in 1972 using TTL called the 832.  It used a 16 word 8-bit programmable ROM made with slide switches.  Another 16 words of RAM was a pair of 7489s TTL 64-bit SRAM.  There were 8 instructions, a serial CPU and a clock of 500 kHz.  We sold several thousand of them as part of a home study course in computer technology.  It worked great and was a great training device.  It is amazing what you can program in just 32 words.

Other kit computers came along in 1974 like the Scelbi-8H, a kit based on Intel’s first 8-bit micro the 8008.  Also in 1974, Radio-Electronics magazine published an article on the Mark-8 a computer also based on the 8008.  Radio-Electronics also published an article on how to make a video terminal with a TV set.  Written by Don Lancaster, the article described a TV Typewriter and the article was eventually expanded into a book by SAMS.

The Intel 8080 arrived in the 1974 time period and that really changed things.  The first real 8080 kit named the Altair 8800 came from an Albuquerque company called MITS.  It was featured on the cover of Popular Electronics magazine and generated massive interest and sales.  It was just a box with CPU and RAM on a bus that became known as the S-100 bus. Programming was in hex with front panel binary lights and switches. A similar machine was also available from IMSAI.  I had one of these in 1976 with a Teletype ASR-33 terminal.  Programming was in BASIC using paper tape I/0.  Slow and primitive but it did work.  The most popular accessory was 4K (yes, 4KB) memory boards.

All hell broke loose after that.  Dozens of computer kits came and went, most using the 8080 or later the Z80.  Some used the Motorola 6800.  The MOS Technology 6502 was also popular (Commodore, KIM and Apple).  Program storage was on an audio cassette using the so-called Tarbell Kansas City standard.  It was primitive but very effective for such low cost.

It was during the 1975-1977 time period all sorts of new computers came along the most notable being the Apple I followed by the Apple II, the Radio Shack TRS-80, and the Commodore PET.  The OS and language was BASIC.  Packaged applications software became available.  I think the Apple II became such a hit mainly because of a fabulous program called VisiCalc.  Remember that?  The first spreadsheet.

Hard disks were not popular because of the cost.  You could get 8” floppy for some machines.  The 5” mini drives came along later and became popular because of their lower cost.  Gary Kildall wrote his famous CP/M operating system for the 8080.  Adam Osborne made the first portable PC with an 8080 and CP/M.

A whole slew of small cheap game computers emerged next.  Remember the Atari 400/800, the Commodore 32, the Sinclair ZX-81 or the Texas Instruments 99/4?

Magazines, computer shows and clubs drove the whole movement.  Some of the magazines were Byte, Kilobaud, Creative Computing, Dr. Dobbs Journal, and Interface Age.  I wrote for Interface Age for a while.  The very first computer show was put together by John Dilks in 1976 in Atlantic City, NJ.  MITS put on an Altair conference in Albuquerque in 1976 as well.  Jim Warren created the first West Coast Computer Faire in 1977. I went to all three shows.   Local clubs sprang up everywhere.  And computer stores became a new retail outlet.

I lost track of all the S-100 bus PCs that were made.  All are history today.  Intel finally announced the 8088 and 8086 16-bit processors in the late 1970s.  The IBM PC came along in August of 1981.  I got one in December.  The IBM PC changed everything.  Suddenly it was no long a hobby business.  And the rest, as they say, is history.

One final point.  The computer business really developed on the east coast thanks to IBM, DEC, Burroughs, Univac, Data General, and other big mainframe and minicomputer computer manufacturers.  Most semiconductor development went on in Silicon Valley, obviously, but much of the computer business itself was elsewhere, less one forgets.  I guess that’s history for you.  Everyone involved has their own view of it.  The truth lies in many places in many forms.

As for Heathkit, that is another whole story I will tell you next time.

Discuss this Blog Entry 27

on Dec 17, 2013

Absolutely right Lou.

Now to me, the first true personal computer was the PDP-11/03. No switches or lights on the front, a resident monitor program handled your commands, all you had to do was attach a Teletype or some kind of keyboard/screen, and away you went. It had a truly superb instruction set and programming capabilities.
Intel would have done well to copy the PDP-11 architecture in their microprocessors. Had they done so the computing world would be very different today.

W.

on Dec 17, 2013

It seems like this is more of an "Everywhere other than Silicon Valley" rather than "East Coast" history.

I wouldn't have been sad if you'd given a whole paragraph to the single-board computer that launched many of us into the active world of programming and creating computer-controlled circuitry: The MOS Technologies Kim-1. Even though there were other "full" microcomputer systems at the time (the Apple ][ came on the scene in 1977 and the single-board Apple preceded the Kim-1) it was the first microcomputer designed to enable engineers (and future engineers) to get a full system up and running with the addition of a power supply. (Many of us made our own power supplies, too!) The apple which it contended with required a lot more be added before you could do _anything_ with it (and the IBM PC was worse, because it came after 5 years of the Apple ]['s supremacy and you still had to pay extra for a monitor and display card before you could use it...and then it wasn't more (for thousands of bucks) than an Apple ][. Hardly big news...

The 6502 was developed by a team which departed Motorola because they had different ideas about how to organize the register set and deal with indexing and indirect addressing. The 6800 was no slouch, either, though, and deserves more than the passing mention it received. Motorola developed it in Illinois, which is more "East Coast" than those 8080 kits out of Albaquirky! Admittedly, not so many kits appeared with 6800s as their basis, but the S100 bus really became a 6800 system pretty quickly.

There were plenty of other microprocessors besides the 4004/8008, many of which impacted the hobbyist world as well as the professional world. Many of them did not originate, or were not made really useful, in Silicon Valley...and some of them are still living a very useful life today, despite the flood of microcomputers-on-an-8-pin-dip today. I understand that the article had to be limited to fit, but there is so much more that deserves mention.

on Dec 18, 2013

In the 1970s I worked as a manufacturer's representative for MOS Technologies. It was tough getting firms to consider using the 6502 processor since there was no second source or SDK (the KIM, Keyboard Interactive Monitor, came late to the party). I still have a fully-functional KIM on my bench.

on Dec 18, 2013

I'd like to add a short note on the Ohio Scientific Superboard SBC's, for nostalgia's sake. In 1979, these 6502 based SBC's were the least expensive computers having a built in BASIC. They also had a keyboard on the computer PCB. If you jumped for the optional 8k of ram (4k was standard), you could get an early version of Microsoft 8k BASIC.

Ohio Scientific went on to produce micro based business systems and was ultimately acquired by M/A-Com.

on Dec 18, 2013

In college, I learned on an Ohio Scientific Challenger with an add-on box (built by the professor and his grad students?). It knew BASIC (which we had to learn by self-study) and a 6502 assembler. External storage was by audio tape but I believe that they hard floppy drive(s?). When I got out of school and got a job, I played the waiting game to buy an Atari 800 with its tech manuals and assembly interpreter in hopes of doing some programming and extension board design.

on Dec 18, 2013

Really? I thought MC832 logic chip.
But an NRI "832" in 1972? Edu toy, no cpu, lame even then.
http://www.oldcomputermuseum.com/nri_832.html
"NOVA - First widely available 16 bit minicomputer manufactured by Data General. This was the first design (in 1968) to use the 74181." Also used in PDP-11 et al.
You neglect Motorola, National Semiconductor, Texas Instruments, Fairchild, all originated in, yup, Texas. Key to all computing you mention was size increase and price decrease of dynamic RAM pioneered by Intel 1103 in California. First COSMAC and Altair models had 256 Bytes of RAM. I bthink the KIM had 1KB of RAM. No RAM = no programs and no data. Apple2, CBM, TRS80 sucked up the PMOS 4 voltage supply (type 4027) 4096x1 DRAM chip market in late 70's. Ditto "no programs and no data" until storage manufacturers e.g. Shugart Associates (now Seagate) provided low cost floppy and hard disk storage (ST506 and SCSI). (I remember the 5MB hard drives originally refrigerator size; PDP11/03 drives were a bit smaller.)

on Dec 18, 2013

I have always enjoyed Ken Polsson's History of Personal Computers. It is comprehensive based solely on product announcements. It starts in 1947 with the transistor and works up from there. He also offers separate timelines that break out only IBM announcements and timelines that break out only Apple announcements. His personal computer timeline runs over 250 pages: http://www.islandnet.com/~kpolsson/comphist/

on Dec 18, 2013

Lou Frenzel is right, the hobby (or personal) computer movement did not start in Silicon Valley. Silicon Valley had not started when people were thinking about personal computers. High school science "nerds" in the 1950s often designed digital projects of one kind or another. Tic-tac-toe machines were built by many, including myself in ninth grade. In those days, I used old pinball machine relays. My senior science project in 1959 was EMAG3, a 3200-tube computer; transistors were expensive. It had wired logic to play a legal game of checkers. EMAG3 got me to the National Science Fair and an MIT scholarship. See a picture at www.diacad.com and newsclip: http://www.youtube.com/watch?v=s9QM5pwmFTM

on Dec 18, 2013

You are all forgetting the one guy who started the entire thing!
Wayne Pickette who my husband worked with at INTEL from 1970 to 1975, had purchased a PDP 8/S from DEC in 1967. He then defined a serial PDP /8S using the 74181 and some logic. In early 1968, he tried to interest Fairchild but they called him crazy. He worked at IBM in 1968 on project Winchester where he used his logic skills to define a portion of that project and subsequently spot an error which resulted in IBM offering to pay for his education. Because of his actions, Art Rock had come to meet him. Art Rock recommended to IBM to hire him. Art Rock told Dr. Noyce about him. Dr. Noyce called him, he appeared at INTEL Jan 2, 1970. He met Dr. Noyce and Dr. Hoff then was put to work immediately. He did not fill out an employment application till April of 1970 hence his official April starting date. To back up the story his social security record shows INTEL paid him the exact same amount of money in 1970 and 1971. I worked at Fairchild and my husband worked at INTEL in Ted Hoff's group. He also verifies that Wayne was there in January 1970. The real story has never been told. But consider these facts:
Wayne Pickette was the only INTEL employee not covered by INTEL's patent agreement. Dr. Noyce allowed that!
Wayne alone demonstrated the 4004 in Las Vegas in November 1971.
Wayne declined to sign the INTEL patent application fearing from a threat he had received in 1967 relating to a patent office action he had recently submitted.
Wayne Pickette was mixed up in the entire basis of the micro-computer/PC development, then, this hero faded into the sunset not to receive his glory.
But guess who Ted Hoff called in 1991 when the INTEL patent had trouble!

on Dec 18, 2013

I still have a MITS Altair 8800 manual on my bookshelf. My wife doesn't understand why I need to keep books like that :-) The company I was working for at the time (about 5 jobs ago) used the Altair 8800 for a couple of years as a development tool for satellite navigation (TRANSIT) receiver software development. I recall having to debug a flaky memory card that turned out not to have any bus receivers on the 10 LSBs of the address bus, but instead connected the S100 bus directly to 32 RAM chips on the board. I called the company that made the board and gave them a piece of my mind about that, and the response was along the lines of "What do you expect for $250?" I think there were a couple of memory boards in the machine like that. I also have a vague recollection of designing a high-speed paper tape reader interface board for the machine. Reloading software through the ASR-33 was too time-consuming with forgetful memory boards. Thanks for the article and the trip down memory lane, Lou.

on Dec 18, 2013

If you really want to look at the real beginnings of computing you wont go far looking up the efforts of Tommy Flowers and Bletchley Park ( niether east or west coast USA try Milton Keynes UK)
Steve

on Dec 18, 2013

Check out Tracy Kidder's book "Soul of a New Machine" for a great read about Data General in Massachusetts. No wonder he won a Pulitzer Prize for the book. His description of a logic analyzer as a camera is brilliant!

on Dec 20, 2013

Some key East Coast (Boston area) items that have not been mentioned that are responsible for the computing environments of today, include:

The New England Computer Society (NECS) was existing at least prior to 1977, a technical home computer group that met at MITRE (and later at BBN) in Bedford, MA. (diacad above was a member)

The Boston Computer Society (BCS), while having very small beginnings in 1977, came to be the largest computer user group with many SIGs, and had influence world-wide.

Visicalc was developed by Dan Bricklin at Harvard and expanded at Software Arts, and was probably the single program responsible for small computer acceptance for business.

BDS-C was created in 1979 by Leor Zolman at MIT, and I believe was the main reason we use C in computers today.

In 1984, I was the original author of the Phoenix PC, XT and AT BIOSes as a consultant to Phoenix Software/Technologies (Norwood, MA), which enabled the consolidation of the entire PC clone industry (and effectively eliminated all other competition except for Apple) at that time. Even today, the only new platform is the tablets' ARM processors.

The Internet was created from the work of BBN, clearly the infrastructure of today's computer environment.

Finally, probably the most ubiquitous software app today was created in 2004 at Harvard by Mark Zuckerberg (Facebook for those not in the know).

on Dec 18, 2013

If we are talking strictky PC, defined as having a Micro, then I don't think we have any option but to say it stated with the Intel 4004 in Silicon valley, but maybe the article proves that this definition is too narrow.
Two other variations come to mind from the early 70s. One was a desktop (calculator style) computer that ran an APL interpreter. I can't remember who made it, and I never actually saw one, but it was definitely a precursor of things at least five years down the road. The other is the series of desktop computers HP made that ran intepreted basic (including GPIB, as the primary market was enginerring instrument control).
As another comment said, the IBM PC was not groundbreaking technology (a lot of stuff from big blue wasn't) It was barely faster than the Apple ][. Its segmented memory architecture and 16 bit registers gave it a little advantage, but the data bus was still 8 bits and the initial memory offering was 64K. The most significant thing it did was give legitimacy to the industry that had previously been considered nothing but a hobby. I remember a company that I worked for had one department that was doing WP and some accounting on a networked set of CP/M machines. I tried to sell management on the value of PCs so I brought an IBM in and demonstrated VisiCalc. Even then it didn't go over well. One VP for finance thought it might be useful for a home budget. Five years later he was doing all his departments budgeting and investing on one! And later the whole oganization ran on a multichip 8086 based Linux "mainframe" using PCs for terminals.

on Dec 18, 2013

As an alternative to the S100 bus, a company called The Digital Group also produced a kit form personal computer based on the Z80. It had separate memory address and data busses and I/O address and data busses. The CPU board was located in a center slot in the backplane with memory on one side and I/O on the other. One of the advantages was that its busses were actively terminated, which greatly increased the performance and reliability of the computer. They also offered as part of the kit, a video/keyboard interface board that could drive a video monitor and interface with a ASCII keyboard.

The Altair 8800 initially only offered a serial interface board that required the user to have access to a teletype to interface with the Altair 8800.

For those who do not know, or remember, MITS, the manufacturer of the Altair 8800, was originally known as "Micro Instrumentation and Telemetry Systems. They made kit form test equipment. Before they started making the Altair 8800 they sold a desktop programmable scientific calculator kit called the 7400. It even had an external program memory expansion unit.

One of the greatest problems with the Altair 8800 was that their power supply was just a step down transformer with a bridge rectifier and filter section. Each plug in board had one or more linear voltage regulators to provide the onboard +5VDC. The transformer provided in the kit was under rated and as users added more expansion boards, the transformer usually failed. Another company offered a replacement transformer, rectifier and filter alternative as a replacement. MITS accused them of being parasites for trying to profit from their product. The company subsequently renamed itself “Parasitic Engineering” to spite MITS.

When MITS ran the two edition article titled "Build your own personal computer" they only expected to sell a few hundred. They had more than that ordered in the first month following the article.

on Dec 18, 2013

in the 70's, I was using the HP 4500? Don't remember the exact name. It was a desktop computer with an integral small crt and keyboard, which ran HP Basic and APL (anyone remember that?) I believe it was developed in Loveland, Colorado. Neither East coast nor West coast,

on Dec 18, 2013

So many memories. My S-100 bus experience was with Mike Simon in Chicago in the late '70s. He sold an S-100 kit at various shows and via ads in Byte magazine. I supplied a card cage for the S-100 mother board to hold the expansion cards in a stable arrangement to take the load off the card slot connectors. The most memorable S-100 project was a control system Mike had contracted to produce for the (hamburger) McDononald's Corporation when their dog & pony show for each of their several dozen franchisees in major cities around the U.S. needed some special mechanical staging and lighting for the Rolling Stones musical entertainment portion of the show. Later on I built IBM 286 and 386 clones for futures traders at the Chicago Mercantile Exchange.

on Dec 18, 2013

The "guys that left Moto" to make the 6502 had some startup issue, since someone stole the M6800 masks to make the first wafers, and cops showed up..rest of story is fuzzy. Anyone know how that all went down? In any case, the 6800 series and the 6500 re-engineered clones were much better at memory management than any intel processors, but IBM picked intel after Moto apparently refused to customize their microcode for IBM. Intel simply put that startup microcode on seperate chip which was reverse engineers quickly to create the IBM PC clone market in Asia. IBM PC share dropped dramatically very quickly with the clone wars. Just as Moto share dropped with its clone wars.

on Dec 19, 2013

That's not quite what happened. Seven guys left Mot and went to MOS Tech. One of them (Mike James), took documents on the chips he was working on at MOT with him to MOS. After Mot filed suit against MOS for patent infringement, etc., the documents were found during discovery.
Allen-Bradley, the principal investor in MOS, decided that they didn't have the stomach for a legal battle with Mot, so they sold their interest back to the founders. Facing an expensive legal battle, MOS was forced to settle the suit. They agreed to pay $200kn in damages, not produce the 6501 (pin compatible with the 6800) and swap patents with Mot.
The principals of MOS were left cash strapped and, shortly after, sold out to Commodore.
Chuck Peddle got frustrated and went back to Mot, where he led the design of the 68000.
Bill Mensch left and started Western Design Center, which designed assorted CMOS and 16-bit variants of the 6502 family. A number of Apple computers used WDC processors.
Don't know what happened to the others..

on Dec 27, 2013

What a time we had in those early days. My first processor was the 8080A in 1974. Its clock input required a 12 volt two phase clock and the data buss would hardly drive 2 TTL loads. We used the Intel 1103 for memory because it was the only low power memory on the market. The 1103 required 18 volt address drivers and the data output was around 200mV. It was no fun interfacing the early 1103. It wasn’t long before Intel provided the 8080 with a 5 volt clock and improved 1103 that didn’t short out pin 9 to 18 from aluminum migration. It seemed like overnight the PC market exploded. The 6502 family was a better design but Intel took the show. About 3 years ago I got to spend some time with Bill Mensch from Western Design Center and not surprising he agreed with me.

on Jan 8, 2014

Thanks for the added inputs.
The "stolen mask" story was told at the time of the first "search" by Moto security people and whomever they called, and my input was from one of the Moto VP's that laughed that they made some copies that still had Moto errors and symbols in the mask layout.
That was long time ago, so stories get retold with embellishments. The sad part was that both companies had better architecture than Intel, who only survived with the IBM PC contract, after first have asked Motorola to simply customize their microcode for a custom solution for IBM...rejected as "standard product only" strategy. Intel put that custom microcode in separate chip, I was told, that was quickly cloned and that started the demise of IBM's PC market share. Then the entire PC industry suffered from the horrific memory management issues with that early Intel architecture. True? Long stories have been told about the work-arounds Intel needed to address more and more memory, while Moto's 68000 series had those problems licked from the start. IBM later partnered with Motorola for its Power series of minicomputers, another story. And Apple used Motorola chips until the consumer volume cost reductions drove them to move to Intel...and now to design their own processors? True? Any comments?

on Dec 19, 2013

COSMAC ELF? RCA1802. Output to a TV. Could play PONG. Radio Electronics cover story. Loved that board...

Newsletter Signup

Please or Register to post comments.

What's Communiqué?

Blogs on topics such as wired and wireless networking.

Contributors

Lou Frenzel

Lou Frenzel is the Communications Technology Editor for Electronic Design Magazine where he writes articles, columns, blogs, technology reports, and online material on the wireless, communications...
Commentaries and Blogs
Guest Blogs
Nov 11, 2014
blog

How to Outsource Your Project to Failure 3

This article will address failure to carefully vet a potential manufacturing or “turnkey” partner and/or failure to transfer sufficient information and requirements to such a partner, a very common problem I have seen again and again with my clients over the years, and have been the shoulder cried upon by several relatives and clients in the past....More
Nov 11, 2014
blog

Transition from the Academe to the Industry Unraveled 1

There have been many arguments here and there about how short-comings of universities and colleges yield engineers with skill sets that do not cater to the demands of the industry. There have been many arguments here and there about an imminent shortage of engineers lacking knowledge in the sciences. There have been many arguments here and there about how the experience and know-how of engineers in the industry may vanish due to the fact that they can’t be passed on because the academic curriculum deviates from it....More
Nov 11, 2014
blog

Small Beginnings 5

About 10 years ago I received a phone call from an acquaintance. He had found a new opportunity selling some sort of investments and he wanted to share it with me in case I was interested. Ken had done fairly well for many years as a contract software developer primarily in the financial services sector. His specialty was writing RPG code. (RPG is often referred to as a write only language.) But he was seeing the handwriting on the wall as the industry moved on to other methods, and saw himself becoming a fossil....More

Sponsored Introduction Continue on to (or wait seconds) ×