The Minicomputer Orphanage

dummy_variable writes: "A friend of mine pointed me to this site, and I thought some people here at Slashdot would probably appreciate it. Quoted from the site: "This is The Minicomputer Orphanage, a place where you can find information on computers from companies no longer in business. It contains sales brochures, reference manuals, and other things. This was started because with the passage of time, these computer systems are just below the Internet event horizon, and there is almost no on-line information on them.""

Comments

Ahhh, memories

[Calle Ballz]

This site, plus www.abandonkeep.com brings a tear to my eye. i remember the commodore system we had in first grade (i actually went to a public school with money for a computer(!)).... We did graphic plotting on it. So proud when i made my first american flag bye plotting red white & blue.......

Re:Ahhh, memories

[Calle Ballz]

nevermind, abandonkeep went down. I am shocked. Anyone know any other good abandon software distributers?

Re:Ahhh, memories

[nachoworld]

Bring back LOGO! What couldn't that turtle do?

---

What a neat source for prior art

[ClarkEvans]

(think patent invalidation)

i can adopt a minicomputer

[SpanishInquisition]

but I can't garantee I won't moleste it.
Thank you

A Golden Age

[Gorobei]

I was in high school when these boxen were being sold. Hardware was so expensive, and was the limiting factor in a computer setup, that people almost didn't consider software as meaningful... a better processor, more RAM address space, upper and lowercase input, color graphics, floppy disk availability, a serial interface: these were the points that drove the purchasing decision.

A hundred companies flowered trying to find the mix that would sell consumers. Within 5 years, the playing field had changed: software drove purchasing decisions... if you didn't have Visicalc, you didn't sell. Almost none of the hardware makers got rich, but many software makers did.

I fondly remember programming the Wang PC with 8K, the Pet vs Apple Flamewars, the ultimate 10 Meg(!) Corvus HardDisk, the highspeed Mountain modems (300 baud,) and cool printers (Epson bi-directional dot-matrix.) We programmers knew intuitively that our job was to wrap the hardware in device-independant drivers. We did it, and commoditized computer manufacture.

Today, our systems rock, but I'm still wistful when I think of the hardware hackers: they were men like us. Who will remember their beautiful designs except us that killed them?

Re:A Golden Age

[localroger]

Today, our systems rock, but I'm still wistful when I think of the hardware hackers: they were men like us. Who will remember their beautiful designs except us that killed them?

I remember them every day, because their designs worked which is something you often can't say of more modern stuff, even when it does run at 1 GHz.

There is also an exponential increase in bloat which usually isn't noticed because the exponent is so very close to 1, just a bit over, and the modern CPUs are so fast. But when you take all the modern lessons in design and try to apply them to a 20 MHz 80186, as one manufacturer I know of has done, and you end up with an interpreter that can only parse a few hundred instructions per second, you might actually think this is normal and somehow reflects a defect in the processor instead of your bloated OO code.

Re:A Golden Age

[RandomPeon]

There's also going to be a tradeoff between complexity, control, and efficiency. You can get the most efficient code possible by writing everything in straight assembly - but no sane person would try it. (I have a friend who claims to know someone who writes Win32 compliant code in straight assembly. Don't believe it....)

In order to design really complex anything, you need more robust and compartmentalized designs - that's what OO gives you. The tradeoff in speed and resource consumption is obvious, but if we didn't make that tradeoff we'd all be stuck with blinding fast BASIC interperters.

the internet horizon

[xeno]

I was reminded about the internet horizon by a recent experience with some ca.1993 network equipment from Intel. Very little documentation and no driver software was available for pre-1995 equipment, and I find the experience rather common with other equipment and software vendors.

In a previous life, I worked with a wide variety of PC hardware, and as a result have _hundreds_ of old (1985-1995) motherboard, drive, NIC, and peripheral manuals boxed up in my basement. I've often thought that I should scan them in (maybe to PDF or OCR->text so that they are text-searchable). However, the rub is that many of these companies are still in business, don't want bother supporting old stuff, and don't want someone else to do it for them, either.

Is there any way I can make these available for people to peruse just as if I had lent them the physical book? They're books, remember. What's the deal with fair use & copying of out-of-print books? Do I have to write an applet viewer that says "I'm sorry, someone else is reading that right now" if there's a simultaneous request for an archival copy of a manual? Better yet, is there a way to legally flip the original and the archival copy so that the physical original is considered the backup for the electronic document?

hmm.

-Jon

Cool. Now where can we find these systems?

[SecretAsianMan]

Now that we have documentation for these systems, where can we find people or organizations that would be willing to donate these types of systems to us for education and historical preservation? I for one would love to have a PDP-8, PDP-11, or a VAX of my own.

--
SecretAsianMan (54.5% Slashdot pure)

Re:Here's one

[unitron]

Did an eBay search and only got one hit for 3 NASA Program Payload Pins, one of which is described as "ASCZ Contel Delta II Built by GE Astro Space", so maybe a couple of years of searching the bowels of General Electric might turn up something.
Doing a google on contel turned up some stuff indicating that contel got bought by GTE or split up between GTE and GE.
Good luck. I fear you will need it.

Re:what happened to Gould

[Guy Harris]

If by "Gould" you're referring to the Gould 32-bit superminicomputers, then

1. your comment was not necessarily off-topic, the moderation it got to that effect nonwithstanding;
2. I don't remember, offhand, when they ditched those machines. I think Gould had bought SEL (Systems Engineering Laboratories, or something such as that), makers of that line of superminis, but the 1960's-to-1990's page in Gould's history stuff on their Web site doesn't mention it - it just mentions that, in the mid-1980's, "The overhaul and restructuring of Gould had proven to be financially disappointing. Many of the newer businesses the company acquired had failed to generate the earnings levels of some of Gould's more steady industrial businesses, which had been divested. As the financial condition of the company deteriorated, various business units were sold to generate cash.", and I suspect the computer business was sold or shut down then.

Damn History Blind PFY's

[mindstorm]

At this point in my Slashdot life, I don't care about karma anymore, so I'll rant:

You damn PFY's don't have any history. Quite a shame.

Heloooow?? Did anyone read the headline if this article? It says MINIcomputer! It's a step between a mainframe and a microcomputer. Are you so clueless to think that a TI99 is a minicomputer?! That thing doesn't have its own tape drive cabinet or the same address space as the computers I saw when I was a kid hanging around DEC on snow days.

I know what a minicomputer is because I was exposed to a VAX when I was 10 years old. I hung out a few times at the DEC plants in Salem, NH and Tewksbury, MA when my stepdad brought me in. I sat at a VT100 and played (well tried to) DnD games and chatted with his coworkers using talk.

At a young age, I got to see what very few kids my age got to see: The Machine Room. The experience was almost religious, spiritual. Brightly lit, white walls, raised flooring, awe-inspiring. The white noise of all the computers and the cooling system drowning out any distracting thought. The machine room was a would on its own. You were surrounded by sheer computing power and was one with it.

This expanded my perspective of computing beyond what sat on my desk or what was plugged in to my TV. It also planted the seed of my inner geekdom. I didn't understand fully all the implications of what I had seen... Until I started using the Internet, and later when I became a sysadmin.

I saw glimpses of the future in 1984 which became the life that we live in now: ethernet, e-mail, the laser printer.

I was there when you PFY's were proto-PFY's.

I really feel sorry for those who missed this milestone in computing. I only caught a glimpse of it. I now wished that I saw more at the time.

Please, take some time to learn the history of computing. The micro was in many ways the foundation of the Internet and the precursor to the client-server model of computing. For many people who experienced mainframes and minis either in an academic setting or just being babysat on a snow day: this is a special time for us. It was a time when computing was a priesthood and the machine room was the cathedral. Please respect our history and our memories.

Re:Here's one

[Soko]

Good god, where did you get THAT thing?? I haven't seen one in over 10 years. When I last had to deal with one of those little critters was just after the Contel Computer Company split off from Contel and was re-named to Versyss. Before that, the company was called CADO systems. Actually, the old CADO machines I serviced were pretty amazing - they could get 4 simultaneous users on an 8085 based system. Yup, a 2MHz, 8 bit CPU. Boy, am I dating myself here.

IIRC, the OS on your mini is some ugly thing called CADOL. The last machine new machine I remember them making was called the Hawk or something - that was an ugly amalgum of Unix and Cadol in the same box, using 2 386 processors. Actually, sounds like you have one of these ugly beasts.

Do a Google search on "Contel Tiger", Versyss, or you may even try contacting these guys - they used to sell those things, and may have a manual or 2 laying around.

Re:Cool. Now where can we find these systems?

[SecretAsianMan]

What's the story on the moderation here? I expressed my willingness to restore and preserve these machines that are important parts of the history of computing. I'm sorry, but I don't see this as trolling of any kind.

--
SecretAsianMan (54.5% Slashdot pure)

Re:Old GE and Honeywell Computers

[Guy Harris]

Is my memory going? Don't know about the Honywell 6180, but I thought I remembered that the GE 635 and GE 645 were 48-bit machines instead of 36-bit as one of the articles says.

Your memory may be going; the item on the multicians.org site about the GE 635 says:

635

General Electric's competitor to the IBM 7094 in the 1965 time frame. Like the 7094, the GE-635 had a 36-bit word, 18-bit addresses, and programmable data channels with direct memory access.

and the item on the 645 says:

645

The computer system that Multics first ran on, produced by General Electric. Derived from the GE-635, a 36-bit word machine with accumulator, quotient register, and 8 index registers like the 7094. Basic execution speed was about 435 KIPS.

The 6180 was also 36-bit (its instruction and register set was largely a superset of the 645's, although I seem to remember the base register set was a little different - instead of pairing two base registers, so that you had 8 base registers but only 4 base pairs, you had 8 double-width base registers; programs that used the base pairs worked on either, but I think programs that tried to use the base registers independently rather than as part of a pair might not have worked).

I know these ain't mini's but....

[Chanc_Gorkon]

They are important to me and I even saw links that led to pages talking about them. Computers that are important to me:

Atari 8 bits (specifically the 800 XL, my First Personal Computer EVER!)

Apple ][ (who didn't play with one of these back in the late 70's and early 80's! I remember learning how to do animation on these....yeah it was blocky, but it was cool)

My Leading Edge 286 machine, my first IBM compatible (I loved this machine! I went through college with it! Leading Edge made great clones...where are they now? Biggest thing I remember about this machine was it was the heaviest computer I ever owned. Totally metal case and the keyboard could break a person's head open.).

Those were my early machine I messed with. Now I play with a Multiprise 2000 (S/390 machine), of course PC's, A couple Ultra 30's and maybe soon a RS/6000 server (for TSM at work). I work IN a computer room. All day, everyday. I see vestiges of computers like those on the minicomputer site in this room. When I was first hired, we had a ES/9000 machine, and several 486 based servers. We have come a long way. By the way, Xerox still uses PDP-11 chips in some of their printer controllers. I run a BIG Xerox 4890 printer and it has a PDP based controller. You thought 8 by 3 filenames were too short? Try 6 by 3's and you cannot change the extension, it must be some type that Xerox uses (FSL, FRM, LGO, JSL, JDL, PDE amongst the types....there are more, but those are the ones I use the most). I believe Xerox even uses some of the Alto like technology in their big printers as well. Their bigger machines use a interface VERY similar to what I saw in Pirates of the Silicon Valley, and other sources depicting those Alto's. Heck I even here that some traffic light manufacturers still use 8085 processors in their controllers. These are slowly changing. Xerox is starting to use Sun machines to run their printers (the above to Ultra 30's are controllers for DP-65's.). Things you wouldn't imagine using PC technology probably do now. Things are changing, but some of the old machines still are useful.

Exactly backward

[localroger]

The simple fact is, as you allude to, writing something in 100% assembly is, for the most part, foolish - you cut down on stability and development time for a small gain in speed.

Well, I write a lot of stuff in assembly or in other higher level languages over assembly, and this is just wrong.

Writing something in 100% assembly increases stability once you've got it working, increases development time (because, as you point out, it's harder to get it working), but trades off for an enormous increase in speed.

IME badly written assembly (that is, quick without a lot of optimization, but by someone who keeps his references straight and doesn't put values in the wrong places) is always noticeably faster than any higher level language no matter how smart the compiler.

I have reverse-engineered some code in the controller I mentioned and it consistently dereferences things it uses in high speed interrupts by 2 or more levels -- that's a 50% or more speed hit no matter how you cut it, and more than that if you judiciously use a few register-level variables. This is the cost of OO languages which must use a lot of dereferencing to achieve their high level of abstraction. A 50% hit may not sound like much until you make the mistake of using it in an instrument with a 6-level interrupt system.