>
>Subject: Re: FPGA VAX update, now DIY TTL computers
> From: Scott Stevens <chenmel at earthlink.net>
> Date: Sat, 12 Nov 2005 19:27:33 -0500
> To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org>
>
>On Sat, 12 Nov 2005 17:16:54 -0500
>Allison <ajp166 at bellatlantic.net> wrote:
>
>> >
>> >Subject: Re: FPGA VAX update, now DIY TTL computers
>> > From: ard at p850ug1.demon.co.uk (Tony Duell)
>> > Date: Sat, 12 Nov 2005 21:33:43 +0000 (GMT)
>> > To: cctalk at classiccmp.org
>> >
>> >>
>> >> On 11/11/2005 at 11:12 PM woodelf wrote:
>> >>
>> >> >PS. As for getting 7400 off scrap boards, if I had the scrap boards I
>> >> >would most likely try to get the scrap item working again. :D
>> >
>> >Me too. I've got piles of obscure boards that I'd like to get going again
>> >sometime. Most have no real practical function, but are part of the
>> >history of computing/electronics.
>> >
>> >About the only thing I'll raid for parts are no-name clone PC
>> >motherboards and cards.. And those are not a good source of standard TTL,
>> >alas.
>> >
>> >>
>> >> There are plenty of junk boards that have no particular useful function.
>> >> I've got a bunch of ISA cards that I salvaged. All were old models of a
>> >> gizmo that appeared to a PC as a monochrome adapter, but which translated
>> >> the CPU writes to video memory to serial codes to a VT-220 type terminal.
>> >
>> >Now that sounds unusual enough to be very interesting. Put it this way,
>> >if I had something like that I'd be trying to track down a dump of that
>> >missing ROM. I would not be raiding it for parts.
>> >
>> >-tony
>>
>> There are tons of mostly unidentifyable boards with usable parts or we
>> know what they are and maybe they even work but heck there's a stack
>> of 100 in the corner with good parts for salvage.
>>
>> Examples of salvage to me XT clones that were never rare, 386 and 486
>> boards with salvageable parts. A sharp TV with a dead red drive from
>> a doitall chip is junk for stripping. Non-salvage items, things I
>> keep and fix like my uVAX ADVICE as it's an in circuit emulator for
>> the uVAX chip so it both maybe the last in existance and unusual.
>>
>> There is junk and not. The junk is there to make the not junk work.
>> One can have too much junk, rarely!
>>
>
>XT and XT clone motherboards are becoming increasingly rare. 286 and 386 motherboards are falling into that class rapidly. 486 motherboards are still around but gradually becoming rare, too.
>
>What I have lying around in big annoying quantities at this point in time is all Pentium and Pentium II motherboards. And actually, most of them are 'lying around' because they've become defective and due to the finer pitch and custom-chip construction are unservicible. All the XT hardware I still have is either working or easily repairable.
>
>Full AT hardware, i.e. motherboards before the 'Baby AT' generation, is close to being 'key' as far as scarcity. The hardware of that era was a magnitude of scale more expensive than the huge fleet 'Baby AT' that came after it. And much of it was built like a Sherman Tank, as compared to the disposable humvee quality of the Baby AT era hardware. Early and slow (less than 16 MHz) full AT '386 motherboards are quite rare now. (aside- yes, in the eyes of the military the Humvee, like the Jeep before it, is a cheap throwaway transport)
>
>And oddly, at this point in time, people like me find ourselves with far more middle-era Sun Microsystems hardware than we can use. Those nice little lunchbox Sparcs and Pizza Box systems are plentiful and you can cart them off in whatever quantity they are available at surplus gear auctions. They used to be the star item at said auctions.
>
>There are historical books now that show all the wonderful architecture that has been destroyed, i.e. the 'Lost Twin Cities' book and video series highlights some marvelous buildings that were demolished in the name of 'progress.' Time rolls on, and what's missed sorely 20 years after is always whatever was knocked down. This also remains the case with old computer gear. Nobody imagined ten years ago that anybody would have any use for a 8 MHz full-AT 80386 motherboard, so nobody kept any.
>
>Ummm. Same as it ever was.
Save for the INboard386 (38616 for AT processor replacement) and the SIIG3000
bick thats a real 5mb, IDE disk and floppy 386/16 powered system I keep as
it's plain handy! It not only does VGA but RS170 video. I do have a small
footprint 386/25 (or 8mhz with turbo jumper off) board I keep as its plain
cute and works too.
Nope nobody kept them. ;)
Allison
>
>Subject: Re: Collectables?
> From: "Chuck Guzis" <cclist at sydex.com>
> Date: Sat, 12 Nov 2005 19:40:32 -0800
> To: cctalk at classiccmp.org
>
>On 11/12/2005 at 10:10 PM Allison wrote:
>
>>Original topic: I'd like to get that Polymorphic to build a small system.
>>
>>I'm thinking Cpro CPUZ, Cpro Disk1A, Interfacer-II and Ram22. That will
>>build a z80, floppy interfaced, 256k with enough parallel IO to do an
>>IDE.
>
>It's yours as a bare board for USD $10 and with 6 nice new gold-plated
>connectors $20. Shipping extra. I seem to recall that Polymophic tied the
>data in and data out together on their boards, but the MB doesn't show
>that.
6??? I thought it was four slots. It's a deal, I'll get you you off line.
>I used to have the IMSAI floppy disk system too, but that wasn't engineered
>nearly as well as the CPU box. One small power supply for each Calcomp
>floppy with some blinkin LED's on a small front panel. I retired the IMSAI
>FDC and Calcomps (the stepper motors eventually both shorted out).
>Replaced the drives with Siemens double-sided units and used Don Tarbell's
>FDC. Much improvement. I used that setup until I retired the system. The
>MITS 4K DRAM boards got replaced with some SSM SRAM (lots of 2102's)
>boards. The CPU was a Z80 board of uncertain vintage.
>
>I suppose the whole mess could be replaced with a Rabbit CPU and a one-chip
>FDC.
Yep but not near as fun or noisy.
>>On a related note I've been doing some 1802 coding and that 1977 NS*
>>box is still running Avocets 1802 assembler under cp/m. Still kicking
>>after 27+ years.
>
>The 1802 was barely capable as a CPU. The only redeeming thing was that it
>was static CMOS and could be run down to a DC clock rate. But that was
>true of the IM6100, too, wasn't it?
Yes the CMOS 6100 and 6120 wer static. The 1802 still has a following.
It's a bizzare part, dumber than an 8048, almost one step down from
a PDP8 but none the less useful and people are still using it! Its
attractive feature was romless with simple front pannel could be
built until I did it with and 8035 and fewer parts. The DMA was nice
but hard to use and killed CPU bandwidth. Yet I have three working
flavors of it and the first was an ELF built off the PE article when
it appeared.
Come to think of it in the homebrew computer hobby the Z80, 6502, 1802,
and maybe the 6809 are the top players to this day.
>Were there any hobbyist systems built around the Fairchild 9440 almost-Nova
>chip?
Oh, the microFlame. Never saw a hobbiest system of one. I think the only
customers that ever saw one were military contractors. Tried to buy one
and got blown of by the fairchild rep. I do have a die in lucite
paperweight for that effort from one of their promotions.
Allison
>
>Subject: Re: Collectables?
> From: "Chuck Guzis" <cclist at sydex.com>
> Date: Sat, 12 Nov 2005 19:18:07 -0800
> To: cctalk at classiccmp.org
>
>On 11/12/2005 at 9:54 PM Allison wrote:
>
>>different from the A version in implmentation. The 8800 (and A) FP was
>>a nightmare of oneshots and was wired to the backplane with a bundle of
>>wires (can you say rinngggg!) plus an wire connection to the CPU card
>>as well.
>
>Miserable white stranded wire. The CPU card connected to the FP as well
>through a molex header with the same cheap white wire. I built my 8800 in
>a weekend, and the wire from the FP to the MB took the longest.
Ah you did the full pluge assembly too. Friday evening to monday morning.
I looked horrible going to work. BUT IT RAN!
>Anything to save a buck here and there. Lots of corners skipped on the
>8800. So much for quality construction of "old stuff'! :)
It was poooor.
>The Integrand has a power supply that one could arc-weld with.
Integrand made one of if not the best crates around for power and
cooling. I have the PS from one that was a spare. Good supply.
Allison
>
>Subject: Re: FPGA VAX update, now DIY TTL computers
> From: Paul Koning <pkoning at equallogic.com>
> Date: Wed, 09 Nov 2005 11:34:10 -0500
> To: cctalk at classiccmp.org
>
>>>>>> "Allison" == Allison <ajp166 at bellatlantic.net> writes:
>
> Allison> After transistors the second evolution in computers was
> Allison> packaging.
>
>True, at least in high end computers. Cray understood packaging -- as
>far back as the CDC 6600 (1964) there is serious magic and serious
>engineering in the packaging.
>
>On the other hand, DEC computers, while nice architecturally, don't
>show anything interesting in the way of packaging.
>
>(A good way to look at it is that you could fit a whole CDC 6000
>series mainframe in a box not much bigger than the PDP-6 general
>registers module, which are roughly contemporary.)
>
> paul
Ah and 1964 is nearly 8 years after TX2.
True for later generations. However CDC6000 was more specialized
packaging where the PDP6 was a whole lot of flipchips and limited
specialized packaging.
Neither are right or wrong only differing goals and approach.
Allison
>
>Subject: Re: FPGA VAX update, now DIY TTL computers
> From: woodelf <bfranchuk at jetnet.ab.ca>
> Date: Fri, 11 Nov 2005 16:14:58 -0700
> To: General Discussion: On-Topic and Off-Topic Posts <cctalk at classiccmp.org>
>
>Chuck Guzis wrote:
>
>>Consider the very old Packard Bell PB250--22 bit words, fewer than 400
>>transistors and 2500 diodes, 63 instructions. Power consumption about 40
>>watts, exclusive of I/O:
>>
>>The trick, of course, is to use bit-serial methods. It seems to me that
>>one could greatly simplify construction of a homebrew machine that way.
>>We're not doing this for speed, right?
>>
>>
>>
>Not really that much, it is registers that hog all the logic gates.
>I guess you would save only about 1/3 of your logic gates.
>
>>Cheers,
>>Chuck
>>
>>
>I got playing around with 74xx logic and the best I can do for my CPU design
>is about 36 chips for a two bit slice. Now who sells 3-input NAND's
>and Gated
>J/K F/F's cheap like they did in the 70's? And lots and lots of 14
>pin Sockets!
>
>
74LS10 dip .25 at ea .17 at Q25 |74F10 dip .19 at ea .14 at Q25 |74HCT10 dip .25 at ea .18 at Q25
JDRmicrodevices.com
Even at a buck a chip if your only using 36 of them thats not painful.
Allison
At 18:07 12/11/2005 -0800, you wrote:
>Let me be clear on the IMS backplane--it's not a 12-slot model, but a
>4-slot version. It appears to be factory-engineered this way and a glance
>at the rails in my Altair 8800 appears to say that this was intended for
>incorporation into a MITS box and not the IMS box. Connection is obviously
>the MITS wire-by-wire method, not the IMS edge connector.
>
>I hadn't realized that IMS made add-ons for the MITS boxes, but here you
>have one.
Ah... I hadn't realized that. I assumed it was a 22-slot backplane from an
IMSAI 8080 - I was going to pull apart the IMSAI and check the number...
Both of my 8800's have 12 slots installed, so I could still use the board,
although chances are I wouldn't actually install it (neither Altair is full,
and 16 slots would be pushing it anyway for the 8800's power supply). Let
me know what you want for it, and if reasonable, and nobody else has a better
use we can talk further.
[I suppose with the front-panel, and the right cards (CPU, 64K RAM, disk and
serial I/O in three cards) I could build a "very short" IMSAI with it - it
would even be kinds cute]
Regards,
Dave
--
dave04a (at) Dave Dunfield
dunfield (dot) Firmware development services & tools: www.dunfield.com
com Collector of vintage computing equipment:
http://www.parse.com/~ddunfield/museum/index.html
>
>Subject: Re: Collectables?
> From: "Chuck Guzis" <cclist at sydex.com>
> Date: Sat, 12 Nov 2005 18:46:26 -0800
> To: cctalk at classiccmp.org
>
>On 11/12/2005 at 9:30 PM Allison wrote:
>
>>Two different companies. IMS was a bit later time frame and did produce
>>upgrades for Altair. In fact just about everyone produced upgrades of
>>one form or another. Typically backplanes, power supplies, RAM, CPU and
>>of course FDC were often seen. About the only part not commonly upgraded
>>in an Altair was the front pannel though that could be done as well.
>
>....and if memory serves, for a time IMS also offered trade-in deals for
>owners of MITS 8800 systems. Having used (but not owned) both, I liked the
>IMSAI a lot better as regards quality of construction than the MITS.
>Eventually all of my cards wound up in an Integrand Systems no-blinking
>lights box with a many-slot actively-terminated motherboard. But I still
>have the MITS box.
>
>Cheers,
>Chuck
Original topic: I'd like to get that Polymorphic to build a small system.
I'm thinking Cpro CPUZ, Cpro Disk1A, Interfacer-II and Ram22. That will
build a z80, floppy interfaced, 256k with enough parallel IO to do an
IDE.
Current topic: Yes, IMSAI did offer a trade in.
I'd used both and prefered the IMSAI. IMSAI was better built hands down
but hit the scene about 6months after MITS. That and the PE article was
visibility that IMSAI didn't initally get. That's why when you an IMSAI
it's more likely to be working than an Altair 8800 (plain or A version).
The 8800B was much improved but way later. However I was fast on the
draw and have a very early Altair which I learned a lot about how to
not engineer things. I hate oneshots as a result. Hence the comments
on mods. Mine was so power short that with 12k ram parallel IO and
a 88ACR casette board the +8volt line was really 7.5...barely. I
rewound the transformer so that would run a reasonable complement of
cards. Mine has every mod but the FP as a result. The FP would have
been next if I didn't GO NS* as the switches from toggling in the boots
were getting really tired. That speaks for why after two years and a
few months when the North*Star box was available I was right on it.
On a related note I've been doing some 1802 coding and that 1977 NS*
box is still running Avocets 1802 assembler under cp/m. Still kicking
after 27+ years.
Allison
