Altair at 50: Remembering the first Personal Computer | Hacker News Reader

Altair at 50: Remembering the first Personal Computer

goto10retro.com

111 pointsrbanffy60 comments

Wow ... the man who invented it, Ed Roberts, had quite the life:

  • Air Force enlisted

  • Air Force comissioned

  • Electrical Engineer

  • Computer inventor

  • "Gentleman" farmer

  • Medical doctor

https://en.wikipedia.org/wiki/Ed_Roberts_(computer_engineer)

Bill Gates goes into this in his autobiography, source code. Paints a very positive picture of him.

I was completely distracted by the magazine cover. The calculator being advertised there, and the wording, caught my attention.

The magazine issued is available here [0].

Apparently, it really was wiring up an entire calculator.

> ... Start assembly by installing and soldering into place the fixed resistors. Then proceed to installing the three electrolytic capacitors, the diodes, and the transistors, taking care to observe proper polarity and basing. Mount the transistors close to the board’s surface...

[0] https://archive.org/details/197501PopularElectronics

Until microcomputer boom of the 70's and 80's there was the calculator boom of the early 70's. The availability of the microprocessor made small calculators widely available for the first time and they were very popular among the type of people that probably would've read Hacker News had it existed.

What caught my eye was the headline about CCDs as successors to video camera tubes. Charge-coupled devices being, of course, the sensor technology now used by essentially all of the billions of digital cameras on the planet.

Hasn’t basically everyone switch from CCD to CMOS sensors these days?

I don’t know the technical difference other than what each stands for.

Oops, indeed. I got the two mixed up.

It’s kind of funny CCDs had a full product life cycle since that magazine cover.

They see a plenty of use in professional and scientific contexts, so it’s not like their life is over. Indeed the largest digital camera sensor ever, the 3.2 gigapixel imager installed in the soon-to-be-online Vera Rubin Observatory, is CCD. CMOS won out in consumer products because of lower cost and power consumption.

[1] https://en.m.wikipedia.org/wiki/Vera_C._Rubin_Observatory

The Altair didn’t survive for very long, but essentially being responsible for the creation of Microsoft is a pretty Big Deal.

Bill Gates: "Popular Electronics" magazine cover of January 1975 change my life"

Paul Allen write the simulator on Harvard PDP-10, Bill Gates write main code of Basic, Monte Davidoff write Math package. They coded day and night during 2 months.

Micro-Soft was born.

Similarly it inspired the start of Apple!

Apple got a couple things right - the II had a keyboard, TV output with graphics, and plug-and-play slots (no fiddling with jumpers and dip switches like s-100 and PCs): each card had a well defined space for IO and ROM.

Note that the Apple I was a kit computer that didn't come with anything other than the board and the chips. Not even a case! It's not really like Jobs and Woz had the brilliant idea that nobody had thought of before to include the accessories, it's just that the prices had come down to the point where a person wouldn't balk at buying a preassembled computer with a keyboard, display and a couple of floppy drives.

> It's not really like Jobs and Woz had the brilliant idea

They understood that, if the computer had a keyboard and a professional-looking case, a lot more people would buy one.

Jobs certainly understood that, but Woz?

https://www.apple2history.org/history/ah04/:

”Jobs thought the cigar boxes that sat on the … desktops during Homebrew meetings were as elegant as fly traps. The angular, blue and black sheet-metal case that housed Processor Technology’s Sol struck him as clumsy and industrial … A plastic case was generally considered a needless expense compared to the cheaper and more pliable sheet metal. Hobbyists, so the arguments went, didn’t care as much for appearance as they did for substance. Jobs wanted to model the case for the Apple after those Hewlett-Packard used for its calculators. He admired their sleek, fresh lines, their hardy finish, and the way they looked at home on a table or desk.”

> Jobs certainly understood that, but Woz?

This is why we need to surround ourselves with people we don't always agree with.

From Paul Allen's Idea Man autobiography, he wrote that he had written an Intel 8008 emulator/assembler on PDP-10 for their Traf-O-Data company.

So adapting the Intel 8008 emulator/assembler to the Intel 8080 was easy, compared to starting from scratch (and wondering if it was even possible that writing an 8080 cross-assembler/emulator would work).

Monte Davidoff has written a math package for PDP-8 so he was hired to write the math package for their BASIC for $400 + more in consulting fees.

Both the Intel 8080 and Motorola 6800 were introduced at $360 - many references to the IBM 360 mainframe computers as the inspiration for this price.

Ed Roberts was able to negotiate thousand Intel 8080 CPUs at $75 (vs $360 list price). A retro computing website mentions that these CPUs were less than perfect somehow, though still functional.

Back to Paul Allen's book, the MITS Altair was originally introduced with a whole 256 bytes of memory. To increase profits, the included memory card was ditched and you had to buy a memory card ($176 or 1K or $264 for 4K - amazon.com sells DDR5 64GB for US$150-$200 today)

    When Intel and Texas Instruments ran short of memory chips, Ed turned to an off-brand called Signetics, whose chips were hopelessly flaky. I started getting calls, less friendly this time, from people who’d invested in 4K of memory and still couldn’t load BASIC. I’d see Bill Yates tearing his hair out in the engineering department over defective Signetics cards. As David Bunnell and Eddie Currie later acknowledged in PC Magazine, “. . . the probability of getting a 4K memory board to work when assembled from [an Altair] kit was remote. And the likelihood that it would continue to work would easily have been rated zero.”

and the fact that MITS couldn't keep up with demand gave birth (IMO) to the 3rd party (licensed and unofficial) peripherals ecosystem.

Altair didn’t survive for very long,

The company sold itself to Pertec for what was a very good amount of money back then.

In HN terms, it was a successful unicorn bro exit.

and if I remember it correctly Roberts used the money to buy a peanut farm? Is that equivalent to a tech bro buying the coast of California or an island?

Did he use all the money, or some?

The property records are probably online. Go nuts.

My point was, buying a peanut farm != spending every penny you owned on the peanut farm. It simply could have been what he desired, enjoyed. He could have socked away way more.

The comment I replied to alluded to said peanut farm being (forgive me) peanuts compared to a coastline.

My school library ( 6th - 8th grades ) had this magazine, and they had a 9-week class on programming in BASIC using a 110 bps teletype connected to an HP2000C that was shared by several school districts. That was my start in all of this. I didn't get my own computer until the C64 price dropped to $200 in 1983.

Nothing beats the sound of a 110 bps teletype. I loved the humming.

Cost the equivalent of $2359 back then, right now you can get a few on eBay for around $5000. Or you can get a mini replica kit for $150. [1]

[1] https://altairmini.com/en/home

Lest you worry you (or your parents) should have stashed away a few Altairs, that $397 in a tax-advantaged retirement account investing in the S&P 500 would be worth more than $70k today.

My dad has an IMSAI 8080 which is a neat piece of hardware. It still works, though a few LEDs failed.

Unfortunately for him, in university one of his professors advised him not to go into computers for a career...

in 1977 or so my dad (a college professor) advised me not to go into computers

a couple of decades later (i hadn't listened, and had been working for high-tech Silly Valley for quite a while) he apologized :-)

I was torn between music and engineering. I would be poorer, but the parties would have been much better.

Olivetti P101 (1965)

https://www.fondazioneadrianolivetti.it/en/p101-cecam/

It's an interesting one. Some people call it a Personal Computer others call it a calculator. However you classify it, it was certainly a interesting and important stepping stone and it's a shame that a lot of European innovations are often forgotten or skimmed over:

https://en.wikipedia.org/wiki/Programma_101

There seem to be many (!) emulators for it.

https://www.s2js.com/altair/

http://altair32.classiccmp.org

https://github.com/simh/simh/blob/master/ALTAIR/altair.txt

https://github.com/njbigwig/AltairEmulator

https://github.com/gloveboxes/AltairEverywhere

etc.

Of course you can run CP/M on it, but then it works like any other CP/M machine (and probably less conveniently than CP/M emulators which are integrated with the underlying host filesystem.)

The article didn't mention the Mark-8 "kit" featured in the July 1974 issue of Radio-Electronics [1]. My dad built the Mark-8 that year and showed me how he could play the Star-Spangled Banner on the AM radio by taking advantage of the RFI.

[1] https://en.wikipedia.org/wiki/Mark-8

For those interested in Ed Roberts, refer to Robert Cringely's "Triumph of the Nerds," in particular Volume 1.

The Kenbak 1 is the first personal computer

https://www.computerhistory.org/revolution/personal-computer...

Nonsense! It was the LGP-30!

(It all depends on the definition or what a personal computer is)

I cannot find it but I think for about a decade the music cart rotation at WREK was powered by a student built automation system on an Altair. Don't remember any faults in the years I was around it.

(I was a music director at WREK in the early '80s)

Wow! Did you take any photos?

Cool I always thought the first PCs were like the TRS or Commodore 64 computers.

There is some controversy - you could actually use a TRS-80, an Apple II or a PET right after taking it out of the box.

I think that, if we define a personal computer as a machine that is designed for a single interactive user, the LGP-30 would be a good candidate. It was not, however, a home computer.

For me, a personal computer needs more than switches and LEDs as its UI. With a serial port, a terminal, and a monitor program in ROM, the Altair would qualify.

TRS-80, AII and the PET came later. You were right on the peripherals. Instead of a serial port, I'd set a TV output with a keyboard as an input, and the Altair would get far more sales.

The people I know who put together Altairs in 1975 used ASR-33 teletypes as terminals to run BASIC.

Teletype and all, the Altair looked and worked like an cut-down entry-level version of the minis that were popular in engineering and science.

Not as powerful as a PDP-8, but less than a tenth of the price.

It was the perfect aspirational project for the electronics hobbyist community of the time.

The fact that you could barely do anything with it wasn't the point. It was a real computer you could set up at home and use without time restrictions or hourly billing.

The S-100 bus market turned into a preview of the PC market. S-100 systems soon sprouted real terminals, floppy drives, and workable memory, and began to appear in the offices of accountants and other non-tech professionals.

The IBM PC probably wouldn't have happened without it. It normalised relatively affordable computing, and the idea of a third party market of expansion cards on a standard bus.

A TV output means you need to create (and have memory dedicated to) the video. A serial port is much simpler.

A 16x64 character screen like the TRS-80 needed 1k of RAM. IIRC, the Altair didn’t have that much memory out of the box.

The video memory would either be in the computer or the terminal. At least if it were in the computer, you could use it for other stuff.

Did it not have a serial port? Would have assumed that connecting it to a terminal or teletype was the standard thing.

We had an Altair at home, that my father assembled from a kit. That version did not have a serial port. It had only the switches on the front panel. I did succeed in inputting some programs and in having them run. But of course it was appallingly limited.

The serial port was its own separate board. https://www.youtube.com/watch?v=RaYTd3dbXeM

There were two very clever I/O ideas that emerged for the Altair: (1) A radio on top of the housing would pick up a signal, allowing audio output! (2) A cassette tape recorder could be used as an external storage device, though I forget how it interfaced... the serial board, I think.

My first computing experience was with an IMSAI 8080 that class assembled the year before.

It had a keyboard and video board, rather than a terminal. The monitor was open chassis to boot (ah the old days when we didn’t protect children from lethal electricity).

It had a ROM monitor and cassette tape. You had to type in (in hex) a short machine language program into the monitor to load BASIC from a cassette. We simply never turned it off.

I tried ti enter the bootstrap through the front panel once, but I made some mistake, and it didn’t work. It was an awful enough experience I never tried again.

Ah man- the power supply in the IMSAI 8080 was scary enough, plus you had the monitor power supply open. Fun times- the "book" sequence on ours was "fat finger in the paper tape reading software, read the cassette IO software from paper tape, load BASIC from cassette, and good to go."

One of the interesting aspects of the Altair was that it was based on a bus called the S-100 bus. You would have a CPU card and a memory card at least, but everything else was optional. The serial board was separate, and strictly not absolutely necessary to play with the computer, since you could enter simple programs directly from the front panel.

I remember S-100 from when I was a kid. Never was hands-on with that hardware but there were all kinds of ads for cards in Byte magazine and others. Seemed like you could get a card for almost anything in S-100 protocol.

That's right. There are still S-100 enthusiasts who are maintaining and developing S-100 cards, see http://www.s100computers.com/ (does not seem to respond correctly to HTTPS right now).

keep the historical context in mind. There were people who wanted a computer at home and people who wanted to bring home the experience they got with access to powerful mainframe & minicomputers at work or school, so there was both a push to build your own computer and a desire to build IO devices like teletypes. The combo, all-in-one is the real revolution (IMO) that you got with the Apple II or the Sol. The TRS-80s and PETs feel a lot more like early commercialization in comparison. Woz was motivated by showing off what he could create, because that's how he communicated. It makes sense that a keyboard and TV - with graphics - shows off way better, same with being able to program in basic "... for $300 you can build a computer that's so good you can type programs on it and run them..."

"Personal computer" can refer either to a small-form computer, in which case the Olivetti P101 was likely the first in 1965 (the PDP-8/E was also a contender), or to a microprocessor-based computer, in which case the Micral N holds that title in 1973.

The Altair, in 1975, was the first commercially successful personal microcomputer.

The Intel line was fathered by the Datapoint 2200; they implemented the CPU in TTL logic boards.

https://en.m.wikipedia.org/wiki/Datapoint_2200

The 6502 line came out Motorola's failure to "indulge" their employees in a low-cost 6800, thus unintentionally fathering MOS Technologies.

https://en.m.wikipedia.org/wiki/MOS_Technology_6502

The 6502 was extremely inexpensive. The first implementation was the KIM-1, so this is the first on that side.

https://en.m.wikipedia.org/wiki/KIM-1

You might want to look at the Sol-20, Micral N, and Kenbak-1 as well. From 1971, the Kenbak was a _personal_ computer, but not a personal _micro_ computer.

The Wang 2200 looks most like we'd expect a personal computer to look like, but the price range was not home-friendly (~$50k today).

If we draw the line at “home computer” we can give Honeywell the crown, even though I doubt even a single one was actually used as a “kitchen computer”

Commodore had the PET in the late 70's, and the Vic-20 before the 64 in the early 80s.

Apple had the II in the late 70s, and before that was the Altair.

I had one from my father in law. Donated it a local university where they had it displayed in the lobby for quite some time.

Reply on news.ycombinator.com