16 Oct
2007
16 Oct
'07
7:04 p.m.
On 10/15/07, woodelf <bfranchuk at jetnet.ab.ca> wrote:
Tony Duell wrote:
PDP-8s come in many sizes - the Straight-8 was "desk-sized" in that it
was available with a built-in desk. The -8/a is about the size of a
microwave oven (1/2 the size of a PDP-8/e), but was also available
built _into_ a desk) the CPU was in front of your knees, with room to
mount two RX01/RX02 units or an RL01/RL02 where the drawers would
normally go. DECmates are the size of large desktop PCs or smaller
(like the DM III)... various sizes and weights, all 12-bits, 4Kwords
to 32Kwords (to 128Kwords in rare cases)...
Check. I have a PDP8/e on my desk.
I
thought a PDP-8 was desk sized. That reminds me to find
a smaller case for SBC6120 so it can fit on a desk.
I have stuffed in a old wooden box.
At one point, I had one mounted in an external floppy case for an Amiga...
http://penguincentral.com/retrocomputing/sbc6120/
As a newbie with just setting up the PDP-8 and OS/8
what is the
best way to get started? I can download stuff from bitsavers
but what to download.
Besides the OS, there's a few collections of BASIC programs running
around - several games and a few of the typical utility-type programs.
If you want to write programs, you'll need the right programs for
that - PAL assembler, FORTRAN-II, FORTRAN-IV (slow on an SBC-6120
since it doesn't have the optional-for-a-PDP-8/a FPP12 math
co-processor), BASIC, etc. BASIC.SV should be on most OS/8 distros
you find out there. The FORTRAN stuff might take a little hunting and
gathering.
DECUS used to have a large PDP-8 collection, but I don't know that
swaths of it are available anywhere.
-ethan
16 Oct
16 Oct
10:04 p.m.
Ethan Dicks wrote:
Besides the OS, there's a few collections of BASIC
programs running
around - several games and a few of the typical utility-type programs.
If you want to write programs, you'll need the right programs for
that - PAL assembler, FORTRAN-II, FORTRAN-IV (slow on an SBC-6120
since it doesn't have the optional-for-a-PDP-8/a FPP12 math
co-processor), BASIC, etc. BASIC.SV should be on most OS/8 distros
you find out there. The FORTRAN stuff might take a little hunting and
gathering.
I still find it hard to belive that PDP-8/a had a math co-processor.
That does make the PDP-8 the smallest computer with real floating
point I can think of. Ben.
17 Oct
17 Oct
12:22 a.m.
woodelf wrote:
I still find it hard to belive that PDP-8/a had a math
co-processor.
That does make the PDP-8 the smallest computer with real floating
point I can think of. Ben.
For a math copro to be called as such, does it have to integrate itself with
the CPU's instruction set - or does some sort of box of tricks sitting on an
I/O port and accessed through normal CPU instructions count? I'm not sure what
the correct definition is.
What do you mean by 'smallest computer' by the way? Smallest in terms of spec?
I'm pretty sure some of the 1950's stuff had optional floating point hardware
available, although it wouldn't have been physically small :)
cheers
Jules
9:32 a.m.
Jules Richardson wrote:
For a math copro to be called as such, does it have to
integrate itself
with the CPU's instruction set - or does some sort of box of tricks
sitting on an I/O port and accessed through normal CPU instructions
count? I'm not sure what the correct definition is.
It does tricks, but then with only 8 opcodes defined on the PDP-8
what do you expect.
What do you mean by 'smallest computer' by the
way? Smallest in terms of
spec? I'm pretty sure some of the 1950's stuff had optional floating
point hardware available, although it wouldn't have been physically
small :)
True, but then they had longer word lengths than a tiny 12 bits.
You are comparing apples and oranges here - late 50's
and 70's
cheers
Jules
Ben.
10:09 a.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
woodelf wrote:
Jules Richardson wrote:
Northstar (of Northstar Horizon fame) sold an S-100 card that had a TTL
FPU on it. It was a microcoded affair (256 words of 40 bits) that
processed a nibble (BCD digit) per clock (4 MHz I believe). You could
also specify how many digits were in the mantissa (2-14, even # digits
only). Only the four basic functions, +-*/, were supported.
Northstar had a version of BASIC that could use this FPU. As I recall,
you had to specify how many digits of precision you wanted in your
floating point numbers when you ordered BASIC (it wasn't dynamically
specified in the interpreter).
Others sold S-100 cards that used an AMD FPU chip. This too required
poking data bytes and a command, then waiting for the result to be
computed and then pulling out the result bytes. It wasn't somehow
integrated into the instruction set of the host processor.
Finally, I recall seeing an article where somebody took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
For a math copro to be called as such, does it
have to integrate
itself with the CPU's instruction set - or does some sort of box of
tricks sitting on an I/O port and accessed through normal CPU
instructions count? I'm not sure what the correct definition is.
It does tricks, but then with only 8 opcodes defined on the PDP-8
what do you expect.
> What do you mean by 'smallest computer' by the way? Smallest in terms
> of spec? I'm pretty sure some of the 1950's stuff had optional
> floating point hardware available, although it wouldn't have been
> physically small :) 
10:21 a.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On 17 Oct 2007 at 12:09, Jim Battle wrote:
Finally, I recall seeing an article where somebody
took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
I remember that one. It may have been an early issue of Byte. I
also wondered at the time if any speed was gained over just running
plain old x80 code.
Cheers,
Chuck
10:48 a.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
"Chuck Guzis" <cclist at sydex.com> wrote:
On 17 Oct 2007 at 12:09, Jim Battle wrote:
This had a couple of S-100 incarnations, one by an outfit known
as Mini-Micro-Mart. In the 1975 timeframe when you had a few K
of memory in a S-100 system it made at least a little sense.
Later extrapolations used a AMD 9511/9512 chip, as in the
Compupro System Support, and boards by TDL and Northstar. That
was a very expensive chip whenever I saw it for sale :-).
Tim.
Finally, I recall seeing an article where
somebody took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
I remember that one. It may have been an early issue of Byte. I
also wondered at the time if any speed was gained over just running
plain old x80 code. 
11:08 a.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
This had a couple of S-100 incarnations, one by an
outfit known
as Mini-Micro-Mart. In the 1975 timeframe when you had a few K
of memory in a S-100 system it made at least a little sense.
Later extrapolations used a AMD 9511/9512 chip, as in the
Compupro System Support, and boards by TDL and Northstar. That
was a very expensive chip whenever I saw it for sale :-).
Tim.
Didn't Marinchip make one as well?
Peter Wallace
1:32 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
Finally, I recall seeing an article where somebody
took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
Well, at least it's better than using solenoids and photocells (which
would have permitted a wider choice of calculators)
2:44 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On 10/17/07, woodelf <bfranchuk at jetnet.ab.ca> wrote:
Fred Cisin wrote:
Yeah.... my mother had a Selectric typewriter back when we got our
first computer (32K PET). She was, at first, interested in the
solenoid-based typewriter adapter, but I think it turned out to be too
expensive for her. I didn't get a printer on a computer until many
years later, and that was dot matrix.
Besides calculators, I remember an article in Byte to adapt a TI
Speak-n-Spell to a computer. Given that the toys weren't cheap, I
don't know how much it would have saved over an SC-02 Votrax chip, but
certainly the Speak-n-Spell hack was interesting.
I was just a bit bummed because as a PET-owning kid, I was just on the
edge of that stuff - the "mainstream" of the day for those sorts of
projects was _not_ my platform, and I was just learning how to do this
stuff and, though I tried a few of these hacks, my success rate was
substantially under 100% (the typewriter solenoid interface was one of
those - there was not plug-and-play PET version. It was strictly
roll-your-own).
But it was all fun and instructional, even if it wasn't always successful.
-ethan
Well, at least it's better than using
solenoids and photocells (which
would have permitted a wider choice of calculators)
That was saved for I/O
devices. :)
3:06 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
> > Well, at least it's better than using
solenoids and photocells (which
> > would have permitted a wider choice of calculators)
> That was saved for I/O devices. :)
On Wed, 17 Oct 2007, Ethan Dicks wrote:
Yeah.... my mother had a Selectric typewriter back
when we got our
first computer (32K PET). She was, at first, interested in the
solenoid-based typewriter adapter, but I think it turned out to be too
expensive for her. I didn't get a printer on a computer until many
years later, and that was dot matrix.
Too bad, those were FUN!
I never came up with any other way to transfer my manuscript into my
publisher's Merganthaler typesetter.
One of my TRS model 3's had a keyboard that wasn't working well. But, if
I hit a key enough times it would start to work. Put the box of solenoids
on top of the keyboard and dumped a bunch of files, and the keyboard
eventually worked OK.
Besides calculators, I remember an article in Byte to
adapt a TI
Speak-n-Spell to a computer. Given that the toys weren't cheap, I
don't know how much it would have saved over an SC-02 Votrax chip, but
certainly the Speak-n-Spell hack was interesting.
Percom peddled a Speak-n-spell interface.
Speak-n-spell was fairly easy to find for $50; a fully assembled Votrax
device was around $300.
To quote Debee Norling, "A Votrax sounds like a Martian in a tin can".
Speak-n-spell was beautifully clear, but with a finite vocabulary,
6:46 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On Wednesday 17 October 2007 17:09, woodelf wrote:
Fred Cisin wrote:
Certain Selectric adaptations come to mind...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
Well, at least it's better than using
solenoids and photocells (which
would have permitted a wider choice of calculators)
That was saved for I/O devices. :)
12:18 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
I also recall seeing an S100 card using a bipolar 8x8 multiplier some
time back in the 70's. I don't recall the brand of the multiplier
chip (Rockwell, TI?).
Cheers,
Chuck
12:55 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
Chuck Guzis wrote:
I also recall seeing an S100 card using a bipolar 8x8
multiplier some
time back in the 70's. I don't recall the brand of the multiplier
chip (Rockwell, TI?).
Was also a pascal chip set, demoed for the S-100 bus.
I am guessing AMD might be the multiplier chip.I think BYTE
had a blurb on signed 8x8 multiplier for the 6800.
Cheers,
Chuck
Ben.
1:30 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On 17 Oct 2007 at 13:55, woodelf wrote:
Was also a pascal chip set, demoed for the S-100 bus.
I am guessing AMD might be the multiplier chip.I think BYTE
had a blurb on signed 8x8 multiplier for the 6800.
Not to go too far astray, but anyone remember the M4TH (or something
like that) S-100 (IIRC) Forth system?
Anyone have one?
Cheers,
Chuck
2:45 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
From: bfranchuk at jetnet.ab.ca
Chuck Guzis wrote:
I have a S100 board made by N* that is a math board but it
doesn't have a FPU. It has a 2901 and some ROM chips.
It only has the one 2901.
Dwight
_________________________________________________________________
Peek-a-boo FREE Tricks & Treats for You!
http://www.reallivemoms.com?ocid=TXT_TAGHM&loc=us
I also recall seeing an S100 card using a bipolar
8x8 multiplier some
time back in the 70's. I don't recall the brand of the multiplier
chip (Rockwell, TI?).
Was also a pascal chip set, demoed for the S-100 bus.
I am guessing AMD might be the multiplier chip.I think BYTE
had a blurb on signed 8x8 multiplier for the 6800.
Cheers,
Chuck
Ben.
3:05 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
dwight elvey wrote:
I actually had an idea for an arbitrary-precision floating point unit
for a bit-serial processor. Maybe I should implement.
Peace... Sridhar
Was also a
pascal chip set, demoed for the S-100 bus.
I am guessing AMD might be the multiplier chip.I think BYTE
had a blurb on signed 8x8 multiplier for the 6800.
> Cheers,
> Chuck
I have a S100 board made by N* that is a math board but it
doesn't have a FPU. It has a 2901 and some ROM chips.
It only has the one 2901.
3:25 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On 10/17/07, dwight elvey <dkelvey at hotmail.com> wrote:
I have a S100 board made by N* that is a math board
but it
doesn't have a FPU. It has a 2901 and some ROM chips.
It only has the one 2901.
Hmm... that sounds interesting - a 4-bit ALU math co-pro? Obviously
it must be running many machine cycles to be digesting numbers a
nybble at a time.
It's not a general-purpose thing, but the Atari vector game hardware
(Battlezone, Star Wars...) has an 8-bit main CPU and a 16-bit
2901-based "math box". It's been mentioned on the list before, but
essentially, the 8-bit CPU feeds shaps to the 16-bit math
co-processor, which does transforms to calculate the 2D vector
endpoints to render the shapes as they tumble and rotate.
-ethan
6:38 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
dwight elvey wrote (in response to a response to a message I posted):
I have a S100 board made by N* that is a math board
but it
doesn't have a FPU. It has a 2901 and some ROM chips.
It only has the one 2901.
Dwight
That is the same board that I was talking that kicked off this
mini-thread. However, reading back, I wrote it a funny way:
Northstar (of Northstar Horizon fame) sold an S-100
card that had
a TTL FPU on it. It was a microcoded affair (256 words of 40 bits) ...
I meant to say the board implemented a FPU in TTL, not that there was a
monolithic TTL FPU chip of some kind.
12:27 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On Wed, 17 Oct 2007, Chuck Guzis wrote:
On 17 Oct 2007 at 12:09, Jim Battle wrote:
IIRC, it was in the .01-centennial issue of Dr. Dobb's. I have a copy of
it somewhere, but there's no way I could find it without a lot of
effort.
--
roger ivie
rivie at ridgenet.net
Finally, I recall seeing an article where
somebody took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
I remember that one. It may have been an early issue of Byte. I
also wondered at the time if any speed was gained over just running
plain old x80 code. 
2:29 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
Roger Ivie wrote:
On Wed, 17 Oct 2007, Chuck Guzis wrote:
The book "How to Design, Build & Program Your Own _Advanced_ Working Computer
System" (1981/TAB Books) has an entire chapter devoted to this, mentioning
a couple of chips.
Note the underlining (part of the title) of "Advanced" to distinguish it from
the earlier book, which was about designing, building & programming
not-so-advanced working computer systems.
I rather questioned the performance aspect too, esp. with using chips from
pocket calcs, which typically weren't too fast. It all seemed like a hardware
hacker's solution to something rather than learning some math and doing some
programming.
Fred Cisin wrote:
On 17 Oct 2007 at 12:09, Jim Battle wrote:
IIRC, it was in the .01-centennial issue of Dr. Dobb's. I have a copy of
it somewhere, but there's no way I could find it without a lot of
effort. Finally, I recall seeing an article where
somebody took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
I remember that one. It may have been an early issue of Byte. I
also wondered at the time if any speed was gained over just running
plain old x80 code. Well, at least it's better than using solenoids
and photocells (which
would have permitted a wider choice of calculators)
The book mentions a programming interface to calculators using relays for the
keyboard matrix and calls such a device a "compulator" (only place I've
ever
heard the expression).
6:49 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On Wednesday 17 October 2007 17:29, Brent Hilpert wrote:
I have both of those, along with a bunch of other stuff from TAB books.
That's some of the worst publishing I've ever seen, typos and errors all
over the place. If I'm remembering right they were in one of those "book
club" kind of deals in one of the magazines back when, and they never did
get my first order right, so I never went anywhere with it...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
IIRC, it was
in the .01-centennial issue of Dr. Dobb's. I have a copy of
it somewhere, but there's no way I could find it without a lot of
effort.
The book "How to Design, Build & Program Your Own _Advanced_ Working
Computer System" (1981/TAB Books) has an entire chapter devoted to this,
mentioning a couple of chips.
Note the underlining (part of the title) of "Advanced" to distinguish it
from the earlier book, which was about designing, building & programming
not-so-advanced working computer systems.
6:45 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On Wednesday 17 October 2007 15:27, Roger Ivie wrote:
On Wed, 17 Oct 2007, Chuck Guzis wrote:
Speaking of DDJ, I could still kick myself for passing up on some bound
versions of their early years, which I first saw at a local computer store
one time for around ten bucks each. (Computerland? I forget...) Seeing
them later the price had gone up 3-4 times that high. :-(
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
On 17 Oct 2007 at 12:09, Jim Battle wrote:
IIRC, it was in the .01-centennial issue of Dr. Dobb's. I have a copy of
it somewhere, but there's no way I could find it without a lot of
effort. Finally, I recall seeing an article where
somebody took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
I remember that one. It may have been an early issue of Byte. I
also wondered at the time if any speed was gained over just running
plain old x80 code.
7:28 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On 17 Oct 2007 at 21:45, Roy J. Tellason wrote:
Speaking of DDJ, I could still kick myself for
passing up on some bound
versions of their early years, which I first saw at a local computer store
one time for around ten bucks each. (Computerland? I forget...) Seeing
them later the price had gone up 3-4 times that high. :-(
I gave a Vol. 1, No. 1 DDJ to someone in exchange for taking a bunch
of other stuff that I didn't want to pay dump fees for.
Sigh.
Cheers,
Chuck
8:23 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On Wednesday 17 October 2007 22:28, Chuck Guzis wrote:
On 17 Oct 2007 at 21:45, Roy J. Tellason wrote:
Well, if somebody has one they don't want, I'd sure like to hear from
them...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
Speaking of DDJ, I could still kick myself for
passing up on some bound
versions of their early years, which I first saw at a local computer
store one time for around ten bucks each. (Computerland? I forget...)
Seeing them later the price had gone up 3-4 times that high. :-(
I gave a Vol. 1, No. 1 DDJ to someone in exchange for taking a bunch
of other stuff that I didn't want to pay dump fees for.
Sigh.
Cheers,
Chuck
8:57 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
woodelf wrote:
Roy J. Tellason wrote:
Mail it to me... I better go eat something,
since I can't think of
my grammer today.
PS.I would rather have kilobyte instead.
Well, if somebody has one they don't want,
I'd sure like to hear
from them...
But mail it me. :) 
18 Oct
18 Oct
4:35 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
Speaking of DDJ, I could still kick myself for
passing up on some bound
versions of their early years, which I first saw at a local computer store
one time for around ten bucks each. (Computerland? I forget...) Seeing
them later the price had gone up 3-4 times that high. :-(
They still come up on ebay and VCM from time to time. Occasionally
you can score a volume for $1+shipping, but more likely you'll be
talking between $10 and $15. Also got a few out of Powells.
I think my set is pretty much complete through v14. They're
certainly much easier to find than the same years of BYTE.
Eric
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
17 Oct
17 Oct
2:44 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
At 09:21 AM 10/17/2007, you wrote:
On 17 Oct 2007 at 12:09, Jim Battle wrote:
I have a MiniMicroMart packet scanned with details on an 8bit
interface to a calculator chipset. Is anyone interested?
Grant
Finally, I recall seeing an article where
somebody took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
I remember that one. It may have been an early issue of Byte. I
also wondered at the time if any speed was gained over just running
plain old x80 code.
6:37 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On Wednesday 17 October 2007 13:21, Chuck Guzis wrote:
On 17 Oct 2007 at 12:09, Jim Battle wrote:
It may have been, as it sounds familiar to me too.
Finally, I recall seeing an article where
somebody took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
I remember that one. It may have been an early issue of Byte. I also wondered at the time if any speed was gained
over just running
plain old x80 code.
I guess in those days RAM was considered quite expensive?
I remember an article which I'm also pretty sure was in early Byte that talked
about how there was plenty of address space, and how 1-bit-wide storage
might be useful for some things.
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
10:28 a.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
Jim Battle wrote:
Yep - the number of systems which could have hardware FPU support added as an
option must be pretty large, even discounting the IBM-compatible PC world.
(I've had some NS32016 stuff with NS32081 FPU on the bench the last couple of
weeks, for instance).
> What do
you mean by 'smallest computer' by the way? Smallest in terms
> of spec? I'm pretty sure some of the 1950's stuff had optional
> floating point hardware available, although it wouldn't have been
> physically small :)
Northstar (of Northstar Horizon fame) sold an S-100 card that had a TTL
FPU on it. It was a microcoded affair (256 words of 40 bits) that
processed a nibble (BCD digit) per clock (4 MHz I believe). You could
also specify how many digits were in the mantissa (2-14, even # digits
only). Only the four basic functions, +-*/, were supported. Others sold S-100 cards that used an AMD FPU chip.
This too required
poking data bytes and a command, then waiting for the result to be
computed and then pulling out the result bytes. It wasn't somehow
integrated into the instruction set of the host processor.
I suspect most solutions are (were) like that (add-on rather than integrated).
Finally, I recall seeing an article where somebody
took a pocket
calculator chip and essentially poked simulated keystrokes at it and
then decoded the LED driver output to determine the answer. It was very
slow, though, so all it saved was the space of the floating point
library code.
OK, that really appeals to my appreciation of people coming up with
off-the-wall solutions to problems. :-)
6:24 p.m.
New subject: FPUs for small computers (Was: What to download for a PDP-8)
On 10/17/07, Jim Battle <frustum at pacbell.net> wrote:
> Jules Richardson wrote:
>
>> For a math copro to be called as such, does it have to integrate
>> itself with the CPU's instruction set - or does some sort of box of
>> tricks sitting on an I/O port and accessed through normal CPU
>> instructions count? I'm not sure what the correct definition is.
I would count those in coprocessor category, for sure.
Northstar (of Northstar Horizon fame) sold an S-100
card that had a TTL
FPU on it. It was a microcoded affair (256 words of 40 bits) that
processed a nibble (BCD digit) per clock (4 MHz I believe). You could
also specify how many digits were in the mantissa (2-14, even # digits
only). Only the four basic functions, +-*/, were supported.
I've seen even simpler home-brew (or more accurately lab-brew) S100
and Apple II boards that support only the multiply function.
I recall a 4bit x 4bit -> 8bit unsigned multiply board that was
essentially 256 bytes of PROM and some address decoding. Write a
byte to the board input port and the board would put the product of
the upper and lower nibbles on the output port. (Actually, the board
I'm thinking about had two output ports, one for the product, and one
for the product with upper and lower nibbles reversed. When it takes
an instruction per bit shifted, you could save a bit of time by
shifting in hardware (i.e. reordering the wires.))
I would imagine that some enterprising students did the same thing
with 64K of PROM for 8x8->16 multiply, once PROMs got big enough for
that to be affordable.
I doubt many of these survive, since labs tend to throw things out and
rarely keep documents around for wire wrapped boards.
Eric
2:37 a.m.
On 10/17/07, woodelf <bfranchuk at jetnet.ab.ca> wrote:
Ethan Dicks wrote:
It's kinda worse than that. The FORTRAN-II compiler compiles down to
rather ordinary and expected sequences of PDP-8 machine instructions.
The FORTRAN-IV compiler, AFAIK, really just compiles down to sequences
of FPP-12 instructions, which makes it fast if you have an FPP-12, but
really slow (because of emulation) if you do not.
There's a bunch of register-level information on the FPP-12 in the
1977-1978 edition of the "Small Computer Handbook" (it has a PDP-8/a
on the cover, naturally). I received a copy of that from the local
DEC field service guys back in 1982 when I picked up my first PDP-8.
Unfortunately, I had an -8/L, and in my "n00bness", failed to
appreciate all that was in there. Much later, when I found a 1968
copy of the handbook, much more made sense. In the meantime, I did
read about the FPP-12 and decided that it would be cool to play with,
but I've never run across one.
-ethan
... FORTRAN-IV (slow on an SBC-6120
since it doesn't have the optional-for-a-PDP-8/a FPP12 math
co-processor)...
I still find it hard to belive that PDP-8/a had a math co-processor.
That does make the PDP-8 the smallest computer with real floating
point I can think of. Ben.
6276
days inactive
6277
days old
33 comments
15 participants
participants (15)
-
bfranchuk＠jetnet.ab.ca -
cclist＠sydex.com -
cisin＠xenosoft.com -
dkelvey＠hotmail.com -
ethan.dicks＠gmail.com -
frustum＠pacbell.net -
grant＠stockly.com -
hilpert＠cs.ubc.ca -
julesrichardsonuk＠yahoo.co.uk -
korpela＠ssl.berkeley.edu -
pcw＠mesanet.com -
ploopster＠gmail.com -
rivie＠ridgenet.net -
rtellason＠verizon.net -
shoppa_classiccmp＠trailing-edge.com