18 Dec
2006
18 Dec
'06
10:31 p.m.
http://klabs.org/history/build_agc/
Thanks to Richard Cini for the link!
No, its not a kit from me, but it is really cool! It has good potential to
become a kit! : )
Its an older link, but very cool. :)
Grant
18 Dec
18 Dec
10:47 p.m.
In article <5.2.1.1.0.20061218212454.02e52910 at pop.1and1.com>,
Grant Stockly <grant at stockly.com> writes:
I've seen this before; has anyone on cctalk built one?
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
11:40 p.m.
On 18 Dec 2006 at 23:47, Richard wrote:
I've seen this before; has anyone on cctalk built
one?
Looks like a labor of love.
But, given that the original uses RTL and core, what's the point of
building one in LSTTL? Why not use an FPGA? It'd be no more or less
authentic. Or perhaps just write a simulator to run on your PeeCee?
What's remarkable about the original is the size (2 cu. ft.) and the
modularity and serviceability.
Cheers,
Chuck
19 Dec
19 Dec
9:17 a.m.
I checked with Frank O'Brien, who I work with at our museum (InfoAge Science
Center) here in New Jersey. Frank is a researcher for the Apollo Lunar
Surface Journal and is co-editor of the Apollo Flight Journal. He replied:
"In 2004, there were two people who *independently* reverse engineered the
guidance computer source*, and then wrote working hardware emulators. This
work was then used by the gent who built actual hardware to run the code.
Now, there is a not-very-well-organized effort to create an Apollo Guidance
Computer (AGC) using Field-Programmable Grid Arrays (FPGA's). FPGA's allow
you to create logic designs using high level design software, and the "burn"
the logic into the chip. In a sense, this is the step after breadboarding,
and before a chip gets mass produced.
All the best,
Frank
*Some of the source was unreadable, so it was recreated by playing with
memory checksums until the values worked out. Talk about dedication!"
I hope that's helpful!
- Evan
-----Original Message-----
From: Chuck Guzis [mailto:cclist at sydex.com]
Sent: Tuesday, December 19, 2006 2:40 AM
To: General Discussion: On-Topic and Off-Topic Posts
Subject: Re: Build your own Apollo Guidance Computer
On 18 Dec 2006 at 23:47, Richard wrote:
I've seen this before; has anyone on cctalk built
one?
Looks like a labor of love.
But, given that the original uses RTL and core, what's the point of building
one in LSTTL? Why not use an FPGA? It'd be no more or less authentic. Or
perhaps just write a simulator to run on your PeeCee?
What's remarkable about the original is the size (2 cu. ft.) and the
modularity and serviceability.
Cheers,
Chuck
9:30 a.m.
In article <458726D7.17136.1434D6AA at cclist.sydex.com>,
"Chuck Guzis" <cclist at sydex.com> writes:
But, given that the original uses RTL and core,
what's the point of
building one in LSTTL?
Is it even possible to build with RTL parts anymore?
I guess you'd have to do it PDP-8 style with discrete transistors?
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
10:09 a.m.
On 19 Dec 2006 at 10:30, Richard wrote:
Is it even possible to build with RTL parts anymore?
I guess you'd have to do it PDP-8 style with discrete transistors?
That's definitely an option--doing it with SMT components could even
come very close to the size of the original. There are also
transistor arrays, if close matching is desired.
Given that the original was modular, I wonder why the copy didn't
implement that aspect. Certainly one could have each type of module
on its own PCB. Core would be a different matter (can one even
purchase monolithic sense amps anymore?), but a concession to modern
technology might not hurt too much in that area.
Cheers,
Chuck
3:25 p.m.
Richard wrote:
In article <458726D7.17136.1434D6AA at cclist.sydex.com>,
"Chuck Guzis" <cclist at sydex.com> writes:
Well I have a dozen or so, maybe if everybody tosses in a few.. :)
There are some pages on the mentioned site giving the specs for the/an IC type
the AGC was based on. It's not really said directly, but it kind of implies
that there was only one type of IC used in the AGC: an RTL dual 3-input NOR in
a 10-pin flat-pak.
Let's see... invert the logic to make use of NANDs, an earlier poster
mentioned 40 tubes of 24 5400's, that's 40*24*4=3840 2-in NANDs... I always
kind of liked the idea of making an entire processor out of just one basic
gate/IC type.
But, given that the original uses RTL and core,
what's the point of
building one in LSTTL?
Is it even possible to build with RTL parts anymore?
31 Dec
31 Dec
12:10 p.m.
Brent Hilpert wrote:
Richard wrote:
I guess one tube would make a bit slice logic block.
While I have been thinking of that and the 3840 sockets have been kind of
holding me back! I have been thinking of 3 gate types. 3-input open collector
nands, 1 input nand ( inverter ) and 2 input nand open collector buffers.
In article <458726D7.17136.1434D6AA at
cclist.sydex.com>,
"Chuck Guzis" <cclist at sydex.com> writes:
Well I have a dozen or so, maybe if everybody tosses in a few.. :)
There are some pages on the mentioned site giving the specs for the/an IC type
the AGC was based on. It's not really said directly, but it kind of implies
that there was only one type of IC used in the AGC: an RTL dual 3-input NOR in
a 10-pin flat-pak.
Let's see... invert the logic to make use of NANDs, an earlier poster
mentioned 40 tubes of 24 5400's, that's 40*24*4=3840 2-in NANDs... I always
kind of liked the idea of making an entire processor out of just one basic
gate/IC type. But, given that the original uses RTL and core,
what's the point of
building one in LSTTL?
Is it even possible to build with RTL parts anymore?
12:01 p.m.
Richard wrote:
Is it even possible to build with RTL parts anymore?
I guess you'd have to do it PDP-8 style with discrete transistors?
Well I
don't like the FPGA idea -- TTL/DTL/RTL logic is still
visible logic. What good is FPGA's if the windows gets updated and
the FPGA software is not supported. PALS and GALS I could see used.
Also the orginal logic was FLAT PACK so if I remember right it could
get very dense.
12:16 p.m.
On 31 Dec 2006 at 13:01, woodelf wrote:
Also the orginal logic was FLAT PACK so if I remember
right it could
get very dense.
IIRC, the basic logic block was a 3-input NOR. You could stay with
RTL and simply implement it in SMT discretes, most likely with no
increase in real estate--lots of little bitty PCBs in place in the
flatpacks.
12:49 p.m.
Chuck Guzis wrote:
IIRC, the basic logic block was a 3-input NOR. You
could stay with
RTL and simply implement it in SMT discretes, most likely with no
increase in real estate--lots of little bitty PCBs in place in the
flatpacks.
That means I have to buy a toaster-oven to do my PCB's. :)
Also I want to go to MARS ... a bigger computer is needed - umm 20 bits???
.
2 p.m.
On 31 Dec 2006 at 13:49, woodelf wrote:
That means I have to buy a toaster-oven to do my
PCB's. :)
Also I want to go to MARS ... a bigger computer is needed - umm 20 bits???
That reminds me; the other day I ran across the datasheet for the NEC
uPD7720 DSP (was it the first IC DSP?). 2 instructions--23 bit
instruction word length. Data memory was 16 bits. Very strange
bird...
Cheers,
Chuck
1 Jan
1 Jan
10:47 a.m.
New subject: early DSP
From: "Chuck Guzis" <cclist at
sydex.com>
On 31 Dec 2006 at 13:49, woodelf wrote:
Hi Chuck
OK, I'll bite. What were the two instructions? Most DSP's
are built around a MAC. I would guess the large word length
might be to include immediate coefficients.
I know some early ones like the Intel 2920 ( not to be confused
with AMD's 2900 series parts ) ran as a continuous loop. I'd
guess the 7720 was similar in that it would just loop.
Most of these type were to create inline filters to shape the
band pass going to and from codec's.
Even though the Intel 2920 was just a simple loop, it was able
to be used for something like a spectrum analyzer with just
the coding. Not too bad for something that doesn't even have
conditional jump instructions.
These types of DSP's were obsoleted when Motorola and TI
came out with DPS's that were more like traditional uP's of the
time. The connected a uP with a MAC. The Motorola was more
easily programmed by someone familar with regular uP's while
the TI's had all kinds of restrictions on how different parts of
the memory could be used.
I have a I2920 developement SDK and a few 2920's. Someday,
I'd drag it out and play some with it.
Dwight
_________________________________________________________________
The MSN Entertainment Guide to Golden Globes is here. Get all the scoop.
http://tv.msn.com/tv/globes2007/
That means I have to buy a toaster-oven to do my
PCB's. :)
Also I want to go to MARS ... a bigger computer is needed - umm 20
bits???
That reminds me; the other day I ran across the datasheet for the NEC
uPD7720 DSP (was it the first IC DSP?). 2 instructions--23 bit
instruction word length. Data memory was 16 bits. Very strange
bird...
Cheers,
Chuck
2:16 p.m.
New subject: early DSP
On 1 Jan 2007 at 10:47, dwight elvey wrote:
OK, I'll bite. What were the two instructions?
Most DSP's
are built around a MAC. I would guess the large word length
might be to include immediate coefficients.
Well, the 7720 has 3 or 4 instructions according to the manual (I
don't know who wrote it, but it's very confusing) which seems not to
be able to make up its mind.
Storage is a 23-bit instruction ROM, a 16-bit data ROM and a 16-bit
data RAM. Architecture is basically 1-address, with two selectable
accumulators.
Instruction I/II, OP and RT: Same format, but the RT also performs a
subroutine return at the conclusion. Fields designate ALU function,
source, destination, and address register modification. Instruction
is single-cycle.
Instruction III, JP: Jump; does conditional, unconditional jumps and
subroutine call.
Instruction IV, LD: Immediate load of value to selected
destination.
Now, I'd say that saying that there are only 3 instructions is sort
of cheating, but then, I didn't write the manual.
NEC doesn't call this chip a DSP; but an SPI (signal processing
interface). But it has the DSP earmark of compute-and-accumulate
capability with a 16x16 single-cycle multiplier.
Literature seems to date the chip sometime in 1983. Unfortunately, I
don't have one of the chips.
Cheers,
Chuck
5 Jan
5 Jan
5:40 p.m.
New subject: New find, Edax PDP 11/23 system available.
I've just found an interesting vintage system.
The machine is an Edax 9100 / 70. It contains a PDP 11/23 with MMU and FPP
chips installed
running RT 11 V4, circa 1980.
The Edax 9100 is a rather odd beast. Its a system that attached to a
scanning electron microscope
and used a cryogenic silicon sensor to measure the spectra of emissions from
the microscopes target
and calculated the elemental composition of the sample.
This system consists of three main sections. Each section connects with
fairly long sets of cables, so
each section is designed to be seperated, built onto or into an existing
scanning electron microscope.
Main chassis:
There is the main chassis, a very mice short roll-around 19-inch rack with a
formica top. This holds
the 11/23 minicomputer in a custom backplane. There are a number of Q-bus
slots, and a large
section of custom backplane holding some fancy custom ADC modules and a
video display and
memory subsystem.
The Q-bus section holds the following boards
M8186 KDF11 (revision unknown, probably an early one)
M8044 DE (memory)
M8028 (no idea)
A Sigma Floppy Disk Controller (RX02 emulation? boots as DY:)
A Dialog ST-506 disk controller
M8016 YB (don't know this one either, I used to drive UNIBUS 11's)
A DEC 4 SLU board, no M# on the handles.
There is a Shugart 8-inch FDD, complete with the original shipping cardboard
to protect the heads.
There is a ST-506 drive in the chassis somewhere, I don't know the capacity,
or if this is a 8 or
5.25 inch drive.
There are a number of power supplies for the electronics, as well as a high
voltage supply for the
special silicon sensor that was attached to the SEM. I don't have the
sensor, its still being used on
the SEM with a newer EDAX controller box.
Keyboard / Console unit:
A very retro-looking, nice keyboard with extra function keys, and a row of
alphanumeric LED
display modules for programmable legends over the function keys. All
apparently controlled by
standard RS-232 links to the 11/23. The displays look like old Litronix
DL-1414 modules.
According to the documentation, the standard keyboard section is used to
boot the RT-11 O/S.
The EDAX application was driven by using the special function keys and
programmable key legends
on the LED displays. One function key dumps you back to the RT-11SJ prompt.
Video Display unit:
Uses a portrait-format CRT in a (then) futuristic tilt-swivel metal chassis.
This interfaces to the
special video display system in the main chassis. This monitor would be
placed next to the normal
display graphics of the scanning electron microscope and would display the
spectra of the sample
being tested along with text displays of the settings, elemental
compositions, etc.
This same display is also used to talk to the PDP/11, I'm thinking that the
UART in the display
section (I have full schematics) is used to emulate a dumb terminal with the
ASCII keyboard
section on the Keyboard / Console unit.
Software:
This thing ran RT-11 SJ V4, and uses a modified version of DEC's basic.
Users could write programs
in BASIC and CALL functions that ran all the analog front end and display
systems. Custom analysis
programs could be written and saved to disk. A sort of batch mode would run
factory and user written
programs to study and test samples.
There is whatever is on the disk, and the original factory 8-inch floppys.
There is even the original Dialog DSD ST-506 disk controller utility and
diagnostic disk with the
Dialog D7026A formatting program.
Condition:
This thing has been removed from service, and has been sitting in the room
next to a working electron
microscope in a top notch analysis lab since its been delivered from the
factory in 1980 or thereabouts.
Its never been moved, or stored in a non climate controlled space. The
condition is excellent.
There are full service schematics and some documentation of the software
(like all the calls to run
the front end and display stuff). The I/O addresses for the custom hardware
appear to be well
documented.
The roll-around rack is quite short, maybe just under 3 feet high. With the
display and keyboard sections
the whole thing is a bit smaller than an Apollo DN660, but very similar in
general layout.
I do not have the (Hubble twist lock) power cable, and I've not tried to
powe the thing up.
Availability:
This thing is in Massachusettes.
Right now this thing is sitting in my office at work with scrap paperwork.
There is a paperwork
process I need to go through to take it out of the building, but I suspect
that the disposal fees the
company would have to pay are such that they would be very happy for me to
remove this thing
for them (free to me).
Moving the thing (during the winter) will be a small pain however.
It will fit into the back of a Jeep Wrangler with the back seat removed, so
I can get this thing home
but I cannot place it into heated storage. I can store it in my garage, but
I'll have to rodent proof
the thing (but it will be stored next to my Lotus).
The little roll-around rack is very nice, so I have to admit I might part
this thing out. So if someone
really wants the Q-bus boards and the floppy and hard disks out of it, lets
talk. Better still, if someone
wants a little RT-11 box, and does not care if the DEC logo is nowhere to be
found, this might be
a fun and funky little box to hack around on.
Its really got killer retro looks, the styling of the thing is really
something (photos are possible).
The keyboard / console section appears to have a standard serial interface,
and its a very cool
keyboard, nice feel, very vintage. That could be easily hacked onto
whatever vintage machine you
wish.
As for the CRT display, given that its got a portrait oriented tube,
probably uses some totally strange
video format that is nearly useless without the Edax display hardware.
Oh yes, there is a high voltage inhibit switch, so you can switch off the
1,000 volt bias supply to the
sensor head if your just using the computer. A nice service feature thats
essential if someone wants to
keep this thing intact and use it as a conventional 11/23 box (but full the
fuse for the HV section, just in
case someone turns that switch back on, you don't want 1 kV hanging off the
end of a disconnected
cable)
Trade offers are welcome,
I'd love to have either a working printing terminal (like a LA34) or just a
printer that can take green
bar paper. Serial or parallel interface will do.
Oh yes, if chopped up, parts can be shipped. If you want the whole thing
you need to come and
pick it up in MA.
5:48 p.m.
New subject: New find, Edax PDP 11/23 system available.
It sounds sweet! I'd love to see some pictures!
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
6:52 p.m.
New subject: New find, Edax PDP 11/23 system available.
At 6:48 PM -0700 1/5/07, Richard wrote:
It sounds sweet! I'd love to see some pictures!
I'll second that (both parts)!
Zane
--
| Zane H. Healy | UNIX Systems Administrator |
| healyzh at aracnet.com (primary) | OpenVMS Enthusiast |
| MONK::HEALYZH (DECnet) | Classic Computer Collector |
+----------------------------------+----------------------------+
| Empire of the Petal Throne and Traveller Role Playing, |
| PDP-10 Emulation and Zane's Computer Museum. |
| http://www.aracnet.com/~healyzh/ |
7:02 p.m.
New subject: New find, Edax PDP 11/23 system available.
On 1/5/07, Bob Shannon <bshannon at tiac.net> wrote:
I've just found an interesting vintage system.
The machine is an Edax 9100 / 70. It contains a PDP 11/23 with MMU and FPP
chips installed running RT 11 V4, circa 1980...
Interesting.
Keyboard / Console unit:
A very retro-looking, nice keyboard with extra function keys, and a row of
alphanumeric LED display modules for programmable legends over the
function keys. All apparently controlled by standard RS-232 links to the
11/23. The displays look like old Litronix DL-1414 modules.
Nice.
Video Display unit:
Uses a portrait-format CRT in a (then) futuristic tilt-swivel metal chassis.
This interfaces to the special video display system in the main chassis...
This same display is also used to talk to the PDP/11, I'm thinking that the
UART in the display section (I have full schematics) is used to emulate
a dumb terminal with the ASCII keyboard section on the Keyboard / Console unit.
That's an interesting way to do it. I'm curious how _dumb_ the
emulated terminal is (VT52/VT100, totally dumb glass TTY, etc.)
Portrait displays are quite unusual in vintage gear, and I'm curious
to learn about any examples.
Better still, if someone wants a little RT-11 box, and
does not care if
the DEC logo is nowhere to be found, this might be
a fun and funky little box to hack around on.
I'm in Ohio, so I'll give this a pass to someone who is much closer,
but it sounds cool.
Its really got killer retro looks, the styling of the
thing is really
something (photos are possible).
I'll "third" the request for photos, especially of the keyboard and
display.
-ethan
7:08 p.m.
New subject: New find, Edax PDP 11/23 system available.
> Better still, if someone wants a little RT-11 box,
and does not care if
> the DEC logo is nowhere to be found, this might be
> a fun and funky little box to hack around on.
Hmmm... How *little* is little? I'd like a small desktop PDP-11. I just gave
away a couple of floor models that were too big for the space I have
available.
7:18 p.m.
New subject: New find, Edax PDP 11/23 system available.
Not desktop little.
Its clearly a floor model machine, maybe a bit over 20 inches wide.
Its maybe 3 feet deep. I don't think its nearly as heavy as a Symbolics
box.
----- Original Message -----
From: "David Betz" <dbetz at xlisper.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Friday, January 05, 2007 10:08 PM
Subject: Re: New find, Edax PDP 11/23 system available.
> Better
still, if someone wants a little RT-11 box, and does not care if
> the DEC logo is nowhere to be found, this might be
> a fun and funky little box to hack around on.
Hmmm... How *little* is little? I'd like a small desktop PDP-11. I just
gave away a couple of floor models that were too big for the space I have
available.
7:23 p.m.
New subject: cctalk frappr? (was: New find, Edax PDP 11/23 system available.)
In article <f4eb766f0701051902x58320a46l2621344435e2eea4 at mail.gmail.com>,
"Ethan Dicks" <ethan.dicks at gmail.com> writes:
Speaking of locations... I know there are
PDP8 owners
<http://www.frappr.com/pdp8owners>
PDP11 owner
<http://www.frappr.com/pdp11owners>
frappr maps.
What other vintage computing frappr maps are out there?
I think it would be very interesting to have a frappr map that
represented all of the cctalkers.
(Now if I could only figure out how to make frappr stop moving from
place to place on the map but just *stay* in one place.)
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
Better still,
if someone wants a little RT-11 box, and does not care if
the DEC logo is nowhere to be found, this might be
a fun and funky little box to hack around on.
I'm in Ohio, so I'll give this a pass to someone who is much closer,
but it sounds cool. 
8:11 p.m.
New subject: cctalk frappr? (was: New find, Edax PDP 11/23 system available.)
Rumor has it that Richard may have mentioned these words:
Speaking of locations... I know there are
PDP8 owners
<http://www.frappr.com/pdp8owners>
PDP11 owner
<http://www.frappr.com/pdp11owners>
frappr maps.
What other vintage computing frappr maps are out there?
There's a RS CoCo Frappr:
http://www.frappr.com/radioshackcolorcomputer
Laterz,
Roger "Merch" Merchberger
--
Roger "Merch" Merchberger | "Profile, don't speculate."
SysAdmin, Iceberg Computers | Daniel J. Bernstein
zmerch at 30below.com |
6 Jan
6 Jan
1:15 a.m.
New subject: New find, Edax PDP 11/23 system available.
On Fri, 2007-01-05 at 22:02 -0500, Ethan Dicks wrote:
That's an interesting way to do it. I'm
curious how _dumb_ the
emulated terminal is (VT52/VT100, totally dumb glass TTY, etc.)
Portrait displays are quite unusual in vintage gear, and I'm curious
to learn about any examples.
Have you heard of the Facit "Twist"?
Facit was, I believe, a Swedish company, made calculators, paper tape
punches... some other miscellany, and this excellent terminal.
The terminal when seen from the side was shaped like an L, with the CRT
tube itself standing (in a plastic casing, of course) on the vertical
part of it. You could grab the CRT and twist it 90 degrees. The terminal
would send an escape code to the system, and the system would redraw the
screen with the altered aspect ratio. It was an early terminal to use a
white phosphor, and I believe it was scanned at 65Hz.
On Tuesday I'm going to a museum which has one of these...somewhere... -
I can try to get a picture of it, if you like. Norsk Data computers
supported them natively in the USER-ENVIRONMENT and NOTIS office
package.
A friend of mine who spent time around the "Studio 54" (Named after the
serial number of the NORD-10 they had been given) computer science
student group at the University, walked in and saw two fellow students
hacking away using a brand new Facit Twist donated to them by Norsk
Data.
"Oh, neat, a Twist!", he exclaimed, and immediately walked over and
rotated the CRT. Unfortunately, they were in the kernel debugger, a
decidedly non-screen-oriented program - the terminal sent its escape
code, and (IIRC) the machine immediately crashed.
-Tore :)
1:33 a.m.
New subject: New find, Edax PDP 11/23 system available.
On 1/6/07, Tore Sinding Bekkedal <toresbe at ifi.uio.no> wrote:
On Fri, 2007-01-05 at 22:02 -0500, Ethan Dicks wrote:
Nope.
That's an interesting way to do it. I'm
curious how _dumb_ the
emulated terminal is (VT52/VT100, totally dumb glass TTY, etc.)
Portrait displays are quite unusual in vintage gear, and I'm curious
to learn about any examples.
Have you heard of the Facit "Twist"? Facit was, I believe, a Swedish company, made
calculators, paper tape
punches... some other miscellany, and this excellent terminal.
I've heard of the company, but don't know the range of their products.
I think I associate the name with paper tape punches mostly.
The terminal when seen from the side was shaped like
an L, with the CRT
tube itself standing (in a plastic casing, of course) on the vertical
part of it. You could grab the CRT and twist it 90 degrees. The terminal
would send an escape code to the system, and the system would redraw the
screen with the altered aspect ratio.
Nice. More than just aspect ratio - somehow, hardware or software, it
would have to change how it renders its bits. I don't _think_ you can
arbitrarily exchange horizontal and vertical deflectors on a CRT, but
you could run a matrix transform on the bitmap in memory and change
all the drawing routines to swap axes without too much trouble, I'd
expect.
On Tuesday I'm going to a museum which has one of
these...somewhere... -
I can try to get a picture of it, if you like.
Please do.
A friend of mine who spent time around the
"Studio 54" (Named after the
serial number of the NORD-10 they had been given) computer science
student group at the University, walked in and saw two fellow students
hacking away using a brand new Facit Twist donated to them by Norsk
Data.
"Oh, neat, a Twist!", he exclaimed, and immediately walked over and
rotated the CRT. Unfortunately, they were in the kernel debugger, a
decidedly non-screen-oriented program - the terminal sent its escape
code, and (IIRC) the machine immediately crashed.
Oops!
The closest thing I had to that experience was working where they had
some public machines set up for classes and casual non-work-related
surfing. Most, if not all, of the machines had an HP LCD with a pivot
feature. You turned the screen to portrait mode, and used the Windows
screen preferences to tell the system what the orientation of the
screen was (no slick little sensor like the "twist", unfortunately).
I spotted the "rotate me" sticker on the LCD, did the dance and told
the machine that it was in portrait orientation. I left in a bit of a
hurry and didn't put it back to landscape mode before I left - someone
came and tracked me down because nobody could figure out how to "fix
it". :-/
I guess "modern" machines don't care which edge is up, but besides the
twist, I don't know of any classic machine that can be used both ways.
I know of the AT&T terminal with a portrait view CRT, and plenty of
arcade machines have the CRT on its "side" (Gorf, PacMan, Tempest...),
but once you get through the rather short list of oddball machines,
you are left with a *huge* pile of landscape-oriented systems that
scan from left to right fast, and top to bottom less fast. My mother
had a "full page display" (portrait orientation) for her Mac SE, but
due to driver/firmware version mismatches, I never got it working. I
think it's somewhere in a closet waiting for me to dig it out.
Love to see those pictures.
Thanks,
-ethan
2:18 a.m.
New subject: New find, Edax PDP 11/23 system available.
On Sat, 2007-01-06 at 04:33 -0500, Ethan Dicks wrote:
I'll do my best. No promises, though - not my camera.
-Tore :)
On 1/6/07, Tore Sinding Bekkedal <toresbe at
ifi.uio.no> wrote:
Yes, I believe that is what it did. Another bit of trivia was that it
was apparently very susceptible to burn-in.
The terminal when seen from the side was shaped
like an L, with the CRT
tube itself standing (in a plastic casing, of course) on the vertical
part of it. You could grab the CRT and twist it 90 degrees. The terminal
would send an escape code to the system, and the system would redraw the
screen with the altered aspect ratio.
Nice. More than just aspect ratio - somehow, hardware or software, it
would have to change how it renders its bits. I don't _think_ you can
arbitrarily exchange horizontal and vertical deflectors on a CRT, but
you could run a matrix transform on the bitmap in memory and change
all the drawing routines to swap axes without too much trouble, I'd
expect. On Tuesday
I'm going to a museum which has one of these...somewhere... -
I can try to get a picture of it, if you like.
Please do. 
8 Jan
8 Jan
2:59 p.m.
New subject: Photos of the EDAX PDP-11/23 box have been taken.
I've taken photos of the Edax box. They are very large (~3.6 meg each), I
have a dial-up connection
(out here in the boonies).
Can someone host these for all interested parties?
I managed to dig deeper into this things chassis. The DIALOG controller
does indeed have a half-height
ST-506 hard disk drive attached.
3:20 p.m.
New subject: Photos of the EDAX PDP-11/23 box have been taken.
In article <001701c73378$b51e1c50$0100a8c0 at screamer>,
"Bob Shannon" <bshannon at tiac.net> writes:
I managed to dig deeper into this things chassis. The
DIALOG controller
does indeed have a half-height
ST-506 hard disk drive attached.
Is it really DIALOG? I thought it was DILOG? (no 'A')
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
3:45 p.m.
New subject: Photos of the EDAX PDP-11/23 box have been taken.
Right, DILOG, my bad.
I've got seven good digital photos, each 3.6 mb.
One shows the keyboard and alpha-led console thingy.
Another shows the front of the complete system stacked up.
There is a photo of the rear of the main chassis, with all the I/O
connectors, etc.
One photo shows the EDAX side of the card cage, all built on DEC format
quad SPC style boards down to the metal rib and extractors. There is a lot
of
hardware here, shielded ADC's, several memories for spectra and analysis,
then
the display boards, and the Q-bus interface.
Another photo shows the PDP 11/23 section with the 50 pin FDD cable and
the 34-pin and 26-pin ST506 interface and an assortment of DEC boards.
A hi-res photo of the rear of the main chassis with the door open shows some
serious hacking potential! Just over half the card cage is taken up by the
special EDAX hardware, quad height modules.
The 11/23 sits off to the side at the bottom of the backplane, you can see
the
Q-bus section clearly. Lots of room for more toys, AND the dual height
backplane above the Q-bus is uncommitted, with a ground plane.
Care to add another Q-bus?
The last photo is a close-up of the rear of the disk drive bay, you can see
part
of the back of a full height 8-inch floppy, and a half height 5.25 inch HDD.
----- Original Message -----
From: "Richard" <legalize at xmission.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Monday, January 08, 2007 6:20 PM
Subject: Re: Photos of the EDAX PDP-11/23 box have been taken.
In article <001701c73378$b51e1c50$0100a8c0 at screamer>,
"Bob Shannon" <bshannon at tiac.net> writes:
I managed to dig deeper into this things chassis.
The DIALOG controller
does indeed have a half-height
ST-506 hard disk drive attached.
Is it really DIALOG? I thought it was DILOG? (no 'A')
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
31 Dec
31 Dec
3:31 p.m.
In article <45981727.2040705 at jetnet.ab.ca>,
woodelf <bfranchuk at jetnet.ab.ca> writes:
Richard wrote:
Sense not making, you.
--
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
<http://www.xmission.com/~legalize/book/download/index.html>
Legalize Adulthood! <http://blogs.xmission.com/legalize/>
Is it even possible to build with RTL parts
anymore?
I guess you'd have to do it PDP-8 style with discrete transistors?
Well I
don't like the FPGA idea -- TTL/DTL/RTL logic is still
visible logic. What good is FPGA's if the windows gets updated and
the FPGA software is not supported. PALS and GALS I could see used. 
1 Jan
1 Jan
4:24 a.m.
Richard <legalize at xmission.com> wrote:
In article <45981727.2040705 at jetnet.ab.ca>,
woodelf <bfranchuk at jetnet.ab.ca> writes:
I have hand-programmed PALs all the way from pencil-and-paper logic to
putting the right volts on the right pins and hitting a one-shot to
do the fuse-blowing. No computer necessary at all. It was decades
ago but I did it!
Non-registered PALs can also be reverse-engineered without any
fancy tools (although I'll admit that I usually use a computer to
do so, a couple times I have found all the possibilities for
"what address does this thing decode?" with just some 7493's and a 555.)
Now all that probably could be done for GALs and even FPGA's but
I haven't done that :-). Even most GALS do not have publicly released
programming specs.
I will gladly admit that the "moral equivalent of the visible PAL" is
the guy who built a processor from discrete surface mount parts
where the "programmable logic" was using/not using arrays (of diodes?
most likely) of parts on pluggable PCB's.
Don't have the URL handy and Google isn't helping. End result seemed
to be more compact and functional for homebrew processor purposes than
DIP PAL's, and a lot more visible.
Tim.
Richard wrote:
Sense not making, you. Is it even possible to build with RTL parts
anymore?
I guess you'd have to do it PDP-8 style with discrete transistors?
Well I
don't like the FPGA idea -- TTL/DTL/RTL logic is still
visible logic. What good is FPGA's if the windows gets updated and
the FPGA software is not supported. PALS and GALS I could see used. 
19 Dec
19 Dec
1:23 p.m.
I've seen
this before; has anyone on cctalk built one?
Looks like a labor of love.
But, given that the original uses RTL and core, what's the point of
building one in LSTTL? Why not use an FPGA? It'd be no more or less
authentic. Or perhaps just write a simulator to run on your PeeCee?
What's remarkable about the original is the size (2 cu. ft.) and the
modularity and serviceability.
6328
days inactive
6348
days old
32 comments
17 participants
participants (17)
-
alexandre-listas＠e-secure.com.br -
arcarlini＠iee.org -
bfranchuk＠jetnet.ab.ca -
bshannon＠tiac.net -
cclist＠sydex.com -
dbetz＠xlisper.com -
dkelvey＠hotmail.com -
ethan.dicks＠gmail.com -
evan＠snarc.net -
grant＠stockly.com -
healyzh＠aracnet.com -
hilpert＠cs.ubc.ca -
jwest＠classiccmp.org -
legalize＠xmission.com -
shoppa_classiccmp＠trailing-edge.com -
toresbe＠ifi.uio.no -
zmerch-cctalk＠30below.com