<signal lines. Every board required it's own regulation, which
<could take 20% or more of the board space, as well as being a
<nightmare to keep cool. If you see early pictures of loaded
<IMSAIs, the cover was always off. This was a necessity, the
<heat was too much with the cover on. I had to use a 16" fan to
<keep mine running with 64KB of 2102 based static RAM (not 21L02s
<BTW, they cost more than the fan did).
Some of the batter boxes had put some thought to air flw and this was not
a problem. The Northstar Horizon was good, VECTOR MZ, COMPUPRO and there
was an oufit that made mostly boxes all well cooled.
In the mid to late 70s it was either boxed systems like TRS-80, Apple
or designer systems like S100 or SS50. SS50 was 6800 based 50 pin bus
and tended to be a very different thinking.
Allison
On Apr 14, 23:22, Tony Duell wrote:
> > On Apr 14, 0:40, David Williams wrote:
> > > You can see pictures of it on my web site on the AT&T 6300 page
>
> Having a text-only machine here, I can't see the pictures, alas...
There's a lot of brown/grey corrosion on the top surface near the terminals,
and not a lot else visible on the PSU itself, though it looks like there's some
on the base of the computer case around the PSU.
> > wipe off any excess WD40, though.
>
> I _hate_ WD40... It's far too easy to misuse and causes too many
> problems.
I prefer WD40 to machine oil for electricial connections, but I have to agree
with Tony that it's very easy to misuse. It should be supplied in
micro-syringes, not spray cans, IMNSHO. Or only available on prescription,
perhaps. That's why I said to be sure to wipe off the excess.
And WD40 should be kept away from moving metal parts. It's sometimes useful as
a plastic anti-squeak lubricant, but definitely not for metals. I use
CLP-BreakFree for that :-)
> and light machine oil. Start with (UK) 600 grit, end with 1000 grit, at
> least. I'd clean the screw terminals with dry 600 grit paper and then
> contact cleaner.
I've seen contact cleaner seriously misused too. Some types leave "a light
film" behind -- if you use a little, once. I once had to fix a BBC Micro with
faulty ROM sockets; the owner had repeatedly fed it contact cleaner until it
was swimming in the stuff. ISTR cleaning most of it off with 1,1,1-TCE and
then washing the board in Teepol (industrial grade/laboratory detergent) before
it could be attacked with a soldering iron.
--
Pete Peter Turnbull
Dept. of Computer Science
University of York
OK... I was talking to a Ukranian programmer, who told me that in 1968 the
PC was invented, not far from where he worked in Ukrane. Now, that's 4
years before the microprocessor, but is it possible?
And this guy might be dilusional, he's VERY communist, but then again,
at base, so am I, but with a democratic twist. Anyway, what's the story
behind this?
Thanks,
Tim D. Hotze
Hi tony,
you know where to get the amd assembler for this stuff ?
cheers,
emanuel
----------
> From: Tony Duell <ard(a)p850ug1.demon.co.uk>
> To: Discussion re-collecting of classic computers
<classiccmp(a)u.washington.edu>
> Subject: Re: neat find
> Date: Tuesday, April 14, 1998 3:14 PM
>
> >
> > today at goodwill I found a Advance Micro Device AM2900 Evaluation &
> > Learning Kit in the it's box (very nice design on it) with one
worksheet.
> > the unit was only $5.
>
> The comment 'you lucky beggar' springs to mind !.
>
>
> The AMD 2900 series of chips were essentially a build-it-yourself CPU.
> The main ones were :
>
> The 2901 - a 4 bit ALU + registers. You could cascade these to give you
as
> many bits as you wanted (in multiples of 4). There was also a fast carry
> generator, equivalent to the 74182. Was that the 2902?
>
> The 2903 was an enchanced 2901 AFAIK. I never used it.
>
> Then there were the 2909 and 2911 4-bit microcode sequencers. You
> cascaded those as well to access whatever size control store you wanted.
> Add a bit of jump logic, and write the microcode to control your CPU.
>
> For simpler designes there was the 2910 12 bit sequencer which couldn't
> easily be extended (*). That would seqeunce a 4K control store, and had
> some of the jump control logic built-in.
>
> I've done a bit with these chips, and was reading the data books earlier
> today, actually. Great pity they're no longer made...
>
> (*) PERQ systems used a 2910 as the sequencer on the 4K PERQ CPU board.
> The 16K CPU board also used a 2910 with a '2 bit kludge' to provide the
> extra address lines. The pun on '2 bit' is intentional, and will be
> understood by anybody who's ever written PERQ microcode...
>
> >
> >
>
> -tony
>> Such innovation, but it ended up in vain efforts just cloning
>> IBM mainframes
>> and the like. Sound like a company you know? (Minus the vain).
>
>Several companies: RCA Spectra, XDS Sigma, Amdahl, Fujitsu, Hitachi ...
> Jack "I love MS, MS loves me" Peacock
You're right. I actually have nothing against MS, just it's current
product line and a few of it's business practices. A good company shouldn't
need to um... err.... ahh... lobby (that's it!) anyone.
Tim "Who Care's About MS, except when you need to" Hotze
> Such innovation, but it ended up in vain efforts just cloning
> IBM mainframes
> and the like. Sound like a company you know? (Minus the vain).
Several companies: RCA Spectra, XDS Sigma, Amdahl, Fujitsu, Hitachi ...
Jack "I love MS, MS loves me" Peacock
>Yes it is possible, if you play with the meaning of "PC". If you define
>it to mean a small computer used by a single person, then a low end mini
>computer becomes a PC. In which case we have to go all the way back to
Yes, but it would have to fit in a space that people can have.. say... at
home.
>machines like the Bendix G-15 (late 50s ?). The USSR did not have
>microprocessors before the US.
And the US was not developing an atomic bomb during WWII.
>That's easy to prove, because soviet
>micros all used US designed instruction sets. How could they produce an
>8080 compatible IC before it was released by Intel?
A better question: How could Intel make a 8080 compatible machine before it
was produced by the Soviets? ;-)
Honestly, I think that Soviet computers are, on the whole, a sad story.
Such innovation, but it ended up in vain efforts just cloning IBM mainframes
and the like. Sound like a company you know? (Minus the vain).
Tim D. Hotze
> OK... I was talking to a Ukranian programmer, who told me
> that in 1968 the
> PC was invented, not far from where he worked in Ukrane.
> Now, that's 4
> years before the microprocessor, but is it possible?
Yes it is possible, if you play with the meaning of "PC". If you define
it to mean a small computer used by a single person, then a low end mini
computer becomes a PC. In which case we have to go all the way back to
machines like the Bendix G-15 (late 50s ?). The USSR did not have
microprocessors before the US. That's easy to prove, because soviet
micros all used US designed instruction sets. How could they produce an
8080 compatible IC before it was released by Intel?
Jack Peacock
>It's easy to blame novice users and get rich quick spammers, but I can't
>agree with you. I look on the 'Net as evolution in action. Right now
>we have the feeding frenzy of spammers drowning us in unwanted email.
>The easy, and wrong, solution is to force them to stop. The right way
>is to make it uneconomical for them to send spam. How? I don't really
>know, I just ignore it, sort of like background static noise on a
>shortwave radio.
In general, it is ignored. I know someone, who recently got online. They
own a small company, and were offered 15,000 eMail addresses at .01 per
person. That sounded like such a great deal! So, they sent the e-mail, and
got 200 eMails that bounced, then in 3 months, had recieved 6 orderes for $5
and over 500 eMails about the bad business practices of spamming. They knew
no better, but still...
>But consider, what kind of response rate do spammers get? Now junk
>snail mailers have to actually pay per piece, although at a reduced
>rate. They have to make the junk mail attractive to readers so they can
>get a high enough response rate to justify the mailing. Maybe we need
>the same mechanism for mass commercial postings. In other words, the
>ISP specifies in the terms of service that mass commercial mailings (aka
>spam) are charged at the same mailing rates as the local post office,
>something around 20 cents per item. Now you have a legal means to get
>back if the ISP catches a spammer, because they are liable for the
>contracted costs.
Spam should be illegal. The commercialization of the Internet's what we
needed for a long time: a worldwide information resource. Allowing people
to profit by it (by ads, etc.) gives you more information available, and a
wider user base. Even sci-fi writers didn't envision a global information
resource like ours for another ten to twenty years. Let's face it: Until we
go past the money-stage, for many things to become popular, they seemingly
MUST be commercial.
<> S-100s started out with an Intel 8080, then Zilog Z80. That was
<> pretty much the end of the 8-bit version, although there were
<
<Hmmm...this is sort of like the Creationist version of where S-100 came
<from. Actually, a guy named Ed Roberts at a company called MITS that
<built a computer called the Altair 8800 invented the S-100 bus.
It's the of the form... In the beginning Ed created MITS and intel created
the 8080 and it was ok,... :-)
<> The response time compared quite favorably to contemporary low
<> end DEC PDP-11s, and for a fraction of the cost.
Actually the Alpha MicroSystems was a s100 box in the 1979-80 time frame
that used the WD13 chipset (same as the LSI-11/pdp-11) with slightly
different microcode and it did run with the same or slightly better
performance than the LSI-11/03 it's contemporary.
<> Another problem was the unregulated power supply. Unregulated
<> +8 and +/-16VDC was run over the bus itself, right next to
<> signal lines. Every board required it's own regulation, which
<> could take 20% or more of the board space, as well as being a
<> nightmare to keep cool. If you see early pictures of loaded
<> IMSAIs, the cover was always off. This was a necessity, the
<> heat was too much with the cover on. I had to use a 16" fan to
<> keep mine running with 64KB of 2102 based static RAM (not 21L02s
<> BTW, they cost more than the fan did).
Typical of the earlier boxen. Their cooling path was poorly thought out
at best and looking at them it's was obvious. Usually some fish paper
in the right places did the trick.
<> >As for laissez-faire, I never have believed in it. It makes society
<> >too concerned about money. This is proven when complete crap hardware
<> >is released now, and people don't care because it's good _financially
An aside to this... some of the S100/SS50/trs80/apple... items from the
various third parties were quite the garbage. Some fo the suppliers were
patently ripoff artists where money was taken and nothing shipped.
Reality check. PCs today can be purchased to do real work and expected
to perform. Computers of the late '70s and into the early 80s were often
useless as shipped if they worked and required a fair amount of acumen to
configure, expand and *required* programming skills. The latter due to
add hardware and even create applications to make them useful.
For example to add a hard disk in 1980 (10mb DISCUS system was $3995.00)
and that was the drive and a s100 card. The user would have to set the
jumpers on the card for the correct IO ports to not conflict with their
system. The system BIOS would have to be rewritten to add the hard disk
and the system tracks on the floppies and the hard disks would have to be
rewritten onto the media. This assumes you had the BIOS for your system
and it was CP/M or similar... all others the bet was off. So to add a
hard disk you needed a system that was up and running, sources, could
read the sources supplied...
I often had the problem of the NS* using hard sectored 5.25 and the
supplied media was 8"!!!
Also you had an assembler, editor, debugger and knew 8080/z80 assembler.
When I went from a teltek controller to a different one... do it all over
again as the new one was totally incompatable. Far cry from an IDE disk
and setting the CMOS.
Allison
