-----Original Message-----
From: Richard Erlacher <edick(a)idcomm.com>
To: Discussion re-collecting of classic computers
<classiccmp(a)u.washington.edu>
Date: Wednesday, October 20, 1999 12:33 PM
Subject: Re: OT: how big would it be?
>Why not explore this problem from the standpoint of an FPGA? When you
>finish you'd still have the flexibility of a hand-built device, yet what
you
>learned in the process would potentially be of actual use?
>
(if this was directed at me)
I have to design that kind of crap now, every day. I just got *into*
transistor computers and find it quite interesting to see what games I can
play with old RTL type stuff.
>Why squander your cash and intellectual resources on creating something the
>folks in the '70's electronics industry were striving to avoid when you
>could have the same mental exercise in a productive form that made the
>design and implementation of your architecture the core of your effort
>rather than issues which are no longer relevant, like power consumption,
>packaging, and finding the appropriate materials from which to build your
>device? Signal routing is the one issue which persists from that era of
>yesteryear when a CPU lived in multiple racks, but it's handled, at least
>superficially, by the development software. Sharing flipflop packages or
>gates between two circuits on a backplane with 50 cards of logic in it is
no
>easy matter. What's more, the propagation delays will slow your circuit
>down to cycles in multiple microseconds, while correcting the associated
>routing errors will take multiple days for each one. While it's not
>perfect, the FPGA approach allows you to have these experiences with a
>"virtual" closet-sized backplane with similarly "virtual" cards (modules)
of
>logic which you can design hierarchically and based on your needs, not on
>what the local parts vendor happens to have.
>
I design high speed logic (some basic config cpu cores) now using Xilinx and
even some AMD chips (yes, I use to use the MACH stuff) all with various RISC
chips. I don't want to come home at night and *continue* the same type of
design work (or code). I find it more entertaining to work on an 8/S , not
drop one in a chip [though that might be interesting]. I try to spend my
time on the old systems which is what taught me back in '79 so much about
the new ones.
This unit will be nothing more than a conversation piece (in my office,
hopefully doing some small task) and I hope to have some fun with quick and
dirty logic.
If my PDP-8/S wasn't so *mint* in the rack I'd rip it out, drop it in a 8/E
table top case and drag it into the office to do something fun.
john
>Dick
>
>-----Original Message-----
>From: Mike Cheponis <mac(a)Wireless.Com>
>To: Discussion re-collecting of classic computers
><classiccmp(a)u.washington.edu>
>Date: Wednesday, October 20, 1999 9:35 AM
>Subject: Re: OT: how big would it be?
>
>
>>> > Say someone were to home-build a CPU from scratch using only
individual
>>> > components, no ICs only modern discrete(?) components. How big would
>the
>>> > CPU be? For comparison lets say it would be an 8080 clone. Any
>guesses?
>>
>>It seems to me the Right Answer is to approximate the number of
transistors
>>required. How many transistors did an 8080 have? (Do remember, however,
>>that the transistor count is actually less than you'd need with discrete
>>transistors, because the on-chip transistors can have multpile emitters or
>>gates or whatever in the same device.)
>>
>>As for how "big" it would be (that is, its size), modern discrete
>transistors
>>are available in tiny SOT-23 or even tinier packages. Resistors are
>>available in 0402 and maybe smaller. Line widths on PC boards can perhaps
>>be as small as .002 inches, and they can be many layers, a dozen or more.
>>
>>So, in order to compute the size, I think you'd need to make two
estimates:
>>
>>1) The number of transistors per cubic inch (or cm if you like those
units)
>>
>>2) The number of transistors required.
>>
>>This assumes some packaging/connector allowance is taken into account to
>>estimate the number of transistors per cubic whatever.
>>
>>------
>>
>>IMHO, if you're going to build something from transistors, why not build
>>something that was originally built with transistors? For example, the
>>IBM 1620, 1130, 1401, etc. Or if you really want to get funky, try
>something
>>like transistorizing a tube machine, like ENIAC or JOHNNIAC or Bendix
G-15.
>>-That- could be entertaining...
>>
>>-Mike Cheponis
>>
>
-----Original Message-----
From: Hans Franke <Hans.Franke(a)mch20.sbs.de>
To: Discussion re-collecting of classic computers
<classiccmp(a)u.washington.edu>
Date: Wednesday, October 20, 1999 12:25 PM
Subject: Re: OT: how big would it be?
<snip>
>(And before y'all argue about backdraws, of course I know that
>it is close to imposssible to get the TTL running at 1 MHz inside
>the NMOS specs, but that can be solved by lowering the CPU clock
>and selecting an aprobiate Application :)
>
>I just belive this would give a _great_ display - unlike all
>these dump displays where they put a sack of transistors beside
>a uP and tell you just that they are equivalent.
>
My point exactly. There is no practical application to build this like there
is no practical reason not to junk every old computer around. They make a
great display and are fun to play with. I'd rather have a few hundred Flip
Chips doing something rather than collecting dust as *spares* on a shelf.
john
>Anyway
>Gruss
>H.
>
>--
>Stimm gegen SPAM: http://www.politik-digital.de/spam/de/
>Vote against SPAM: http://www.politik-digital.de/spam/en/
>Votez contre le SPAM: http://www.politik-digital.de/spam/fr/
>Ich denke, also bin ich, also gut
>HRK
>
Hey! That must be the same UART board I have lying about somewhere. It has
a bunch of shift registers in to-5 cans (to save space) and calculates
parity using a JK flipflop.
This thread seems to have started with the notion of even building the
flipflops from discrete transistors and passives. That 4x5-inch board would
grow to the size of a closet door using that thechnology. What's more, the
power would have to be distributed with #16 wire.
Dick
-----Original Message-----
From: CLASSICCMP(a)trailing-edge.com <CLASSICCMP(a)trailing-edge.com>
To: Discussion re-collecting of classic computers
<classiccmp(a)u.washington.edu>
Date: Wednesday, October 20, 1999 9:46 AM
Subject: Re: OT: how big would it be?
>>I have trouble with the notion of the uart filling a 9 x 11 board given
>>that I'm holding one that's occupying 4 x 5 inches in SSI. Yeah, the shift
>>registers would take a bunch of space but I don't see it using anywhere
>>near the amount of real estate suggested.
>
>I'd suggest a electromechanical (or optomechanical) UART instead. You
>know, like in a Teletype :-).
>
>>> A pdp-8 (early) had a pannel roughly 24"x50" with flip chip modules
mostly
>>> transistors and the 4k core was a 10" tall rack section. for rough
>>> comparison. In many respects the 8080 is a far more complex CPU and
would
>>> be significantly bigger. It would also be slow compared to the NMOS
part.
>
>>I suspect you could build a pdp-8 using contemporary layout tools and
discrete
>>technology that, excluding the core stack, was an order of magnitude
smaller.
>
>And repackaging would also save a lot of money: a large part of the cost
>of a Straight-8 is all those gold plated fingers and edge connectors, and
>the backplane wiring. Get rid of that - so that your CPU resides on
>a single (even if large) PC board - and you're way ahead. (Well, way
>ahead if everyone else is still in 1965...)
>
>>> Doing it in ttl or bit slices would still be big, I've done that. using
>>> 2900 parts(ca mid to late '70s) the CPU equivelent was over 100 chips
and
>>> filled 4 10x8" cards.
>
>>That sounds about right; I recall building a PDP-11 clone using 2901/2910
parts
>>as part of an undergraduate CPU architecture course in the same era and
using
>>about the same number of parts.
>
>Of course it's also possible to do it on a single card using SSI and
>MSI TTL, maybe with a few bipolar PROM's. Take a look at the 11/04 CPU or
>the original Nova, for example.
>
>I recall - back in the mid-70's - that Radio Shack sold transistor-based
>logic module kits (PC boards) that could be strung together to make
>things like binary counters, etc. Does anyone else here remember these?
>Or, even better, still have the modules around?
>
>--
> Tim Shoppa Email: shoppa(a)trailing-edge.com
> Trailing Edge Technology WWW: http://www.trailing-edge.com/
> 7328 Bradley Blvd Voice: 301-767-5917
> Bethesda, MD, USA 20817 Fax: 301-767-5927
Why not explore this problem from the standpoint of an FPGA? When you
finish you'd still have the flexibility of a hand-built device, yet what you
learned in the process would potentially be of actual use?
Why squander your cash and intellectual resources on creating something the
folks in the '70's electronics industry were striving to avoid when you
could have the same mental exercise in a productive form that made the
design and implementation of your architecture the core of your effort
rather than issues which are no longer relevant, like power consumption,
packaging, and finding the appropriate materials from which to build your
device? Signal routing is the one issue which persists from that era of
yesteryear when a CPU lived in multiple racks, but it's handled, at least
superficially, by the development software. Sharing flipflop packages or
gates between two circuits on a backplane with 50 cards of logic in it is no
easy matter. What's more, the propagation delays will slow your circuit
down to cycles in multiple microseconds, while correcting the associated
routing errors will take multiple days for each one. While it's not
perfect, the FPGA approach allows you to have these experiences with a
"virtual" closet-sized backplane with similarly "virtual" cards (modules) of
logic which you can design hierarchically and based on your needs, not on
what the local parts vendor happens to have.
Dick
-----Original Message-----
From: Mike Cheponis <mac(a)Wireless.Com>
To: Discussion re-collecting of classic computers
<classiccmp(a)u.washington.edu>
Date: Wednesday, October 20, 1999 9:35 AM
Subject: Re: OT: how big would it be?
>> > Say someone were to home-build a CPU from scratch using only individual
>> > components, no ICs only modern discrete(?) components. How big would
the
>> > CPU be? For comparison lets say it would be an 8080 clone. Any
guesses?
>
>It seems to me the Right Answer is to approximate the number of transistors
>required. How many transistors did an 8080 have? (Do remember, however,
>that the transistor count is actually less than you'd need with discrete
>transistors, because the on-chip transistors can have multpile emitters or
>gates or whatever in the same device.)
>
>As for how "big" it would be (that is, its size), modern discrete
transistors
>are available in tiny SOT-23 or even tinier packages. Resistors are
>available in 0402 and maybe smaller. Line widths on PC boards can perhaps
>be as small as .002 inches, and they can be many layers, a dozen or more.
>
>So, in order to compute the size, I think you'd need to make two estimates:
>
>1) The number of transistors per cubic inch (or cm if you like those units)
>
>2) The number of transistors required.
>
>This assumes some packaging/connector allowance is taken into account to
>estimate the number of transistors per cubic whatever.
>
>------
>
>IMHO, if you're going to build something from transistors, why not build
>something that was originally built with transistors? For example, the
>IBM 1620, 1130, 1401, etc. Or if you really want to get funky, try
something
>like transistorizing a tube machine, like ENIAC or JOHNNIAC or Bendix G-15.
>-That- could be entertaining...
>
>-Mike Cheponis
>
-----Original Message-----
From: allisonp(a)world.std.com <allisonp(a)world.std.com>
To: Discussion re-collecting of classic computers
<classiccmp(a)u.washington.edu>
Date: Wednesday, October 20, 1999 9:49 AM
Subject: Re: OT: how big would it be?
>> BTW: I am building an Intel 4004 or possible an 8008 with flip chip
modules.
>> Will tell you when I start it.
>
>Those parts were in the 3000-6000 transistor bracket. I'd expect an 8008
>in flip chips would be bigger than a straight-8 (that's larger than an
>8S). Why? The 8008 is register rich compared to the PDP-8 series and
>for every bit in a register thats a flip chip (8008 is about 45 bits of
>registers alone and then there are the FFs in the state logic). This
>thing is going to be HUGE. The 4004 would be less ambitious but there are
>still a lot of FFs in that part.
I am going to impliment the instruction set serially... Hmm.. wonder where
I've seen that before?
I have run out of transistor/old computers to restore so I rfeally liked the
8/S and am going to do some fun stuff with a few hundred flip chips.
john
>
>Allison
>
I wrote:
> Tracks per inch 48 TPI both drives
> Tracks per drive TM100-1 40 tracks, TM100-2 80 tracks
I usually read my copy of the ClassicCmp Digest between midnight and 1
o'clock
(when I receive it). And if I wasn't so busy copying this out of the
manual, I
might have read what I was writing. :)
Bill wrote:
> | |
> | |
> \ /
> \ /
> | | < BREAK HERE
> | |
> \ /
> -
Thats just where they broke! I used one of my wife's sewing needles to
pry them out
of the sockets. And it's most likely they broke when I removed the chips
>from the
sockets, but I'm not sure about that.
Dave Dameron wrote:
> The bad chips were all made by TI.
Four of the 5 broken chips are TI 8T26s and the fifth is a TI 74123.
--Doug
====================================================
Doug Coward dcoward(a)pressstart.com (work)
Sr. Software Eng. mranalog(a)home.com (home)
Press Start Inc. http://www.pressstart.com
Sunnyvale,CA
Curator
Analog Computer Museum and History Center
http://www.best.com/~dcoward/analog
====================================================
>I know its off-topic but i figured that since most of the poeple on this
>list work or have worked on the really big stuff you'd know better than
>most others.
>
>Say someone were to home-build a CPU from scratch using only individual
>components, no ICs only modern descrete(?) components. How big would the
>CPU be? For comparison lets say it would be an 8080 clone. Any guesses?
Well, others have guessed at the 8080 Clone, but I'll step in and point
out that if minimal part count is an important feature, then you can
get by with a *lot* less. Especially if you go to a bit-serial
architecture. (Something that's still mentioned in many computer
architecture textbooks, even if it isn't used much anymore!)
How big? I'll venture a guess that a 12-bit serial CPU
could be done on one largish (i.e. a square foot) PC board, using
TO-92's, resistors, capacitors, and diodes. This isn't too far
out of line with many bit-serial designs from the late 1960's
(for example, many bit-serial desktop calculators used a PC board
- or two or three - about this size.)
Total cost? Maybe $1500 in large scale production, including testing
costs.
Of course, you now hook up memory to this CPU. Doing that with
discretes would be a chore!
--
Tim Shoppa Email: shoppa(a)trailing-edge.com
Trailing Edge Technology WWW: http://www.trailing-edge.com/
7328 Bradley Blvd Voice: 301-767-5917
Bethesda, MD, USA 20817 Fax: 301-767-5927
--- Joe <rigdonj(a)intellistar.net> wrote:
> Ethan,
>
> They're in a surplus place near Orlando. They're covered with dust. They
> look like they've been there a LONG time. I have no idea what they want for
> them. Give me an idea of what they're worth.
As usual, I have no idea. If I saw them locally (within a few hours drive),
and priced at $50 each, I'd snap them up in a second. If they were tagged
at $100 each, I might get them. $500 each, I'd leave 'em be. I am afraid
for what it would cost to ship them to Ohio from central Florida. Several
dollars per pound truck freight, no doubt, plus the effort/expense of
palletizing. At least both should be able to go on the same pallet.
Thanks for the tip, though. Generally speaking, my rescue range is 8 hours
drive, but it had better be good to be at the edge. The last road trip I
did was to Dayton for a 15Mb 11/750 plus some goodies.
-ethan
=====
Infinet has been sold. The domain is going away in February.
Please send all replies to
erd(a)iname.com
__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com
Hi Chris. I just picked up a Nova 2 loaded with a Diablo 40 series drive,
graphics unit, Kennedy 9800 tape, and a "network" interface. Along with it
came 6 large boxes filled with manuals (schematics, interfacing, theory,
software manuals, and languages,revision notes,nuclear reactor control
software......).
Also came was 130 pounds (I repeat 130 pounds) of paper tape software
including every possible version of rdos, basic, fortran, algol, test tapes,
(other languages)....[many unopened and not installed]....
If you are looking for any documents or programs I probably have them and
can make the images available to you. I have not powered up the system yet
as I have to test everything first (and am too busy on the PDP-8/S finishing
touches).
john
.
-----Original Message-----
From: Chris Kennedy <chris(a)mainecoon.com>
To: Discussion re-collecting of classic computers
<classiccmp(a)u.washington.edu>
Date: Tuesday, October 19, 1999 9:33 PM
Subject: Diablo 31/33 drives
>I'm looking for a diablo 31 or 33 drive (the 31 was the one with
>the door, the 33 is the one that you have to rip apart to get
>your pack out). I'm also looking for a Decision 3150 controller
>for same.
>
>While I'm wishing for the impossible, I might as well add the
>other (current) items on my wish-list:
>
>EDS mux boards for the Nova bus
>An Eclipse S/200
>An Eclipse S/130
>A paper tape reader. Hell, I'd be happy with an ASR 33 at this point.
>
>And in the only-in-my-dreams category:
>
>An original ("baby") nova or supernova
>An Eclipse S/230
>
>Of course, if it's a 16 bit DG machine, something that plugs into
>same or something that just looks good sitting next to it I'm
>interested. Cash, trades or body parts.
>
>Thanks,
>Chris
>--
>Chris Kennedy
>chris(a)mainecoon.com chris(a)dtiinc.com
>http://www.mainecoon.comhttp://www.dtiinc.com
>
>> Say someone were to home-build a CPU from scratch using only individual
>> components, no ICs only modern descrete(?) components. How big would the
>> CPU be? For comparison lets say it would be an 8080 clone. Any guesses?
>
> One of the vendors of bit-slice components (AMD 2900 or clone, I think)
> offered a board that emulated an 8080. In addition to being faster than
> an 8080, you could of course modify the microcode to add or change
> instructions. IIRC, it looked like the board had about fifty chips.
>
> If you didn't use a bit-slice, you'd have to use separate ALU, memory,
> and shifter chips, and you'd wind up with even more.
>
> The early microprocessor architectures were designed based on minimal
> transistor count for a single-chip implementation. This does not result
> in minimal chip count if you implement the equivalent in 7400-series
> chips. It is quite possible to design useful processors with a lot fewer
> TTL chips.
Er, Eric? He said, no ICs only modern discrete components...
As Eric says, an 8080 clone or similar would not be good to do with TTL chips or
something, but it might work well with discretes.
How big? Depends on the routing technology:
A few trannies to a PCB, standard modules where possible, do anything
complicated on the backplane => as big as a PDB8 (which someone mentioned) - say
a 2 foot cube.
Something more modern, say a multilayer pcb covered with densely packed surface
mount transistors => you might get it onto a board the size of an AT
motherboard, or possibly two such boards (allow for plugs on one to go straight
into sockets on the other, mount the boards side by side so you can place
connectors in arbitrary locations over the board)
Something in between is probably more practical - still pin through hole, single
or possibly double sided, but with modern, denser PCB (or pen-wired) layouts,
and not worrying about using standard modules (unlike DEC, you're not
mass-producing) => it would probably be a rack full of Eurocards. (20 * 5 * 10
inches, roughly). Hmm, perhaps a double rack. 20 * 10 * 10.
I haven't looked at transistor counts or packing densities so these are only
guesses.
My opinion (FWIW) is that while this might be a fun project, doing something
with a machine that was originally designed this way (PDP8 again) might be more
educational.
Philip.
This E-mail message is private and confidential and should only be read
by those to whom it is addressed.
If you are not the intended recipient, you are hereby notified that any
dissemination, distribution, copying, reproduction, modification or
publication of this communication is strictly prohibited. Please delete
the message from your computer and destroy any copies.
This message is not intended to be relied upon by any person without
subsequent written confirmation of its contents. This company therefore
disclaims all responsibility and accepts no liability of any kind which
may arise from any person acting, or refraining from acting, upon the
contents of the message without having had subsequent written
confirmation.
If you have received this communication in error, or if any problems
occur in transmission please notify us immediately by telephone on
+44 (0)2476 425474
