Re: [PiDP-1] Digest for pidp-1@googlegroups.com - 25 updates in 7 topics

31 views
Skip to first unread message

Michael Cheponis

unread,
Dec 30, 2025, 4:39:57 PM12/30/25
to pid...@googlegroups.com
Sorry if this has been previously posted:   https://www.youtube.com/watch?v=novOJ_wEvaU
2025 Holiday Sing-Along with Peter on the pdp-1, Diane Z. providing Good Cheer, and Ken Sumrall leading the 'choir'.

-Mike


On Tue, Dec 30, 2025 at 8:30 AM <pid...@googlegroups.com> wrote:
Randy Merkel <rme...@acm.org>: Dec 29 10:15PM -0800

The console is completed and running the blinky app in all its glory! Lots
of solder bridges, my soldering kung fu is what it once was...
 
Some odds and ends;
 
- Waiting on a 22-30 AWG wire stripper for the controls; I don't trust
myself to strip them by hand ;)
- I tried the "hack" to fix the top right LEDs not sticking out enough.
Kind of a pain to get right. I think I'm going to pick up some dowels and
cut to length, between the PCBA and the ottopanel, and use double sided
tape to see if that works better. Anybody try something similar?
- Has anybody identified good enclosure for the game panels? Something
like a small, flat "blue box"would do the trick, but I guess some old cigar
boxes might do the trick!
- Finally; any plans to make an official type 30 display kit?
 
Thanks again, cheers;
 
-- Randy
Matthias Barthel <maklumatp...@gmail.com>: Dec 30 08:26AM +0100

Randy,
 
Michael build both, a type30 display and cases for the Spacewar-controllers
 
https://hackaday.io/project/203991-dec-precision-crt-display-type-30-reproduction
 
Its also anywhere here in the google group .
 
Matthias
 
'Randy Merkel' via [PiDP-1] <pid...@googlegroups.com> schrieb am Di., 30.
Dez. 2025, 07:15:
 
Bill E <wjegr...@gmail.com>: Dec 29 12:42PM -0800

It add listings and cleans up some code. Here is a sample that shows
interbank coding also with some convenient macros that are predefined for
accessing other memory banks, e.g. farlio() to load the IO register with a
value from any bank.
 
Updated docs also here.
As usual, all in https://github.com/wjenh/pidp1-mods.git
 
I'm now happily writing code that needs multiple banks, so I'm sure I'll
find some bugs.
 
Bill
Matthias Barthel <maklumatp...@gmail.com>: Dec 29 01:46PM -0800

Bill,
 
I'm too stupid to compile AM1 :)
I type "make" in the AM1 directory and get this:
 
bison -o y.tab.c -d parser.y
make: bison: Datei oder Verzeichnis nicht gefunden
make: *** [Makefile:22: y.tab.h] Fehler 127
 
what have i done wrong ?
 
matthias
 
Matthias Barthel <maklumatp...@gmail.com>: Dec 29 01:52PM -0800

so i installed bison on my computer , but:
bison -o y.tab.c -d parser.y
gcc -O0 -g -c am1.c
gcc -O0 -g -c symtab.c
gcc -O0 -g symtabgen.c am1.h symtab.o -o symtabgen
symtabgen permsyms.def
make: symtabgen: Datei oder Verzeichnis nicht gefunden
make: *** [Makefile:13: permsyms.def.c] Fehler 127
 
 
Matthias Barthel schrieb am Montag, 29. Dezember 2025 um 22:46:32 UTC+1:
 
Bill E <wjegr...@gmail.com>: Dec 29 01:57PM -0800

That's strange. It says that the permsyms.def file can't be found. Check
your dir to see if it's there. It's in the repo. It could also be a path
issue, be sure '.' is in your PATH env var so the current directory will be
searched.
Bill
Matthias Barthel <maklumatp...@gmail.com>: Dec 29 02:01PM -0800

Bill,
 
i have a solution that works for me.
i must install 2 tools
sudo apt install bison, flex
 
and in the makefile i added a ./ to every "symtabgen" call - yet i have AM1
in my directory
 
Matthias
 
Milo Velimirović <milovel...@gmail.com>: Dec 29 04:01PM -0600

> gcc -O0 -g -c am1.c
> gcc -O0 -g -c symtab.c
> gcc -O0 -g symtabgen.c am1.h symtab.o -o symtabgen
 
am1.h should pobably be am1.o
 
I have not tried this, it’s just decades of programming in C that’s flagging that as wrong.
 
—Milo
 
Bill E <wjegr...@gmail.com>: Dec 29 02:02PM -0800

I just copied the Tools/AM1 stuff to a dir on my machine, it builds. So,
something in your enviroment, and again I bet it's your PATH setting. You
should see:
~/Foo$ make
bison -o y.tab.c -d parser.y
gcc -O0 -g -c am1.c
gcc -O0 -g -c symtab.c
gcc -O0 -g symtabgen.c am1.h symtab.o -o symtabgen
symtabgen permsyms.def
gcc -O0 -g -c permsyms.def.c
flex -d lexer.l
gcc -O0 -g -c lex.yy.c
gcc -O0 -g -DYYDEBUG=1 -c y.tab.c
gcc -O0 -g -c parsefns.c
gcc -O0 -g -c eval.c
gcc -O0 -g -c maccodegen.c
gcc -O0 -g -c bincodegen.c
gcc -O0 -g -c listcodegen.c
gcc -O0 -g am1.o permsyms.def.o symtab.o lex.yy.o y.tab.o parsefns.o
eval.o maccodegen.o bincodegen.o listcodegen.o -o am1
Peter Long <pl...@insys.com.au>: Dec 29 11:12PM

I just built this under MacOS and it needed a few tweaks - so while ymmv here they are:
 
The ./ in the makefile is up to you - I don't have . in my path - but maybe you do ... but the symtabgen rule needs a tweak.
 
plong@mini AM1 % diff Makefile Makefile.orig
13c13
<     ./symtabgen permsyms.def
---
>     symtabgen permsyms.def
46d45
<     $(CC) symtabgen.c symtab.o -o symtabgen
plong@mini AM1 %
 
plong@mini AM1 % diff parser.y parser.orig.y
59,61c59,61
< int yyerror(const char *errstr);
< void verror(const char *msgP, ...);
< void vwarn(const char *msgP, ...);
---
> void yyerror();
> void verror(char *msgP, ...);
> void vwarn(char *msgP, ...);
906c906
< vwarn(const char *msgP, ...)
---
> vwarn(char *msgP, ...)
921c921
< verror(const char *msgP, ...)
---
> verror(char *msgP, ...)
934,935c934,935
< int
< yyerror(const char *errstr)
---
> void
> yyerror(char *errstr)
940d939
< return 0;
 
Compiling eval.c complained about non-void functions without return() statements but I can live with that !
 
Cheers
 
Peter
 
 
________________________________
From: pid...@googlegroups.com <pid...@googlegroups.com> on behalf of Bill E <wjegr...@gmail.com>
Sent: Tuesday, December 30, 2025 09:02
To: [PiDP-1] <pid...@googlegroups.com>
Subject: [PiDP-1] Re: AM1 update posted to github
 
I just copied the Tools/AM1 stuff to a dir on my machine, it builds. So, something in your enviroment, and again I bet it's your PATH setting. You should see:
~/Foo$ make
bison -o y.tab.c -d parser.y
gcc -O0 -g -c am1.c
gcc -O0 -g -c symtab.c
gcc -O0 -g symtabgen.c am1.h symtab.o -o symtabgen
symtabgen permsyms.def
gcc -O0 -g -c permsyms.def.c
flex -d lexer.l
gcc -O0 -g -c lex.yy.c
gcc -O0 -g -DYYDEBUG=1 -c y.tab.c
gcc -O0 -g -c parsefns.c
gcc -O0 -g -c eval.c
gcc -O0 -g -c maccodegen.c
gcc -O0 -g -c bincodegen.c
gcc -O0 -g -c listcodegen.c
gcc -O0 -g am1.o permsyms.def.o symtab.o lex.yy.o y.tab.o parsefns.o eval.o maccodegen.o bincodegen.o listcodegen.o -o am1
 
--
You received this message because you are subscribed to the Google Groups "[PiDP-1]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-1+un...@googlegroups.com<mailto:pidp-1+unsubscribe@googlegroups.com>.
To view this discussion visit https://groups.google.com/d/msgid/pidp-1/ce59c328-227b-4987-b028-31ae66beee54n%40googlegroups.com<https://groups.google.com/d/msgid/pidp-1/ce59c328-227b-4987-b028-31ae66beee54n%40googlegroups.com?utm_medium=email&utm_source=footer>.
Bill E <wjegr...@gmail.com>: Dec 29 05:18PM -0800

On Monday, December 29, 2025 at 6:12:23 PM UTC-5 pl... wrote:
 
I just built this under MacOS and it needed a few tweaks - so while ymmv
here they are:
 
 
Interesting. Apparently MacOS has some strange ideas about things, at least
its C compiler does.
The yyerror decl is consistent in parser.y and matches what Bison says it
should be, not sure where the int bit is coming from. Online docs say
either is acceptable, though.
The eval.c no return is also bogus. From what I see, the only non-return
cases are if yyerror is called, which never returns, it exits. Of course
the compiler doesn't know that, but I do.
Is it insisting on const, or is that just a personal preference? I've never
bothered with it, too much 'training wheels' aspect for me.
I did add ./ to symtabgen. I never thought about that, I've always (since
Unix early days) had . in my path.
 
Bill
Peter Long <pl...@insys.com.au>: Dec 30 05:23AM

Bill
 
Sorry - there were a couple more things it didn't like , which I forgot to put in the last email so here are all the changes needed to compile without errors & most warnings [those remaining don't matter] on a late model Mac Mini / MacOS ...
 
In am1.c - the getenv declaration, the SIGCHLD name change and the char ch needed to be moved ...
 
The make rule for symtabgen needs to be changed as the original rule fails
 
parser.y needed changes to yyerror and verror args and type
 
lexfuncs.c needed some extra parentheses to remove an ambiguity
 
Compiling eval.c complained about non-void functions without return() statements but I can live with that !
 
Yep - Mac's here to help ;-)
 
 
Peter
 
diff am1.c am1.orig.c
98a99
> char *getenv();
131c132
< signal(SIGCHLD, SIG_DFL); /* special case */
---
> signal(SIGCLD, SIG_DFL); /* special case */
685d685
< char ch;
720a721
> char ch;
plong@mini AM1 %
 
plong@mini AM1 % diff Makefile Makefile.orig
46d45
<     $(CC) symtabgen.c symtab.o -o symtabgen
plong@mini AM1 %
 
plong@mini AM1 % diff parser.y parser.orig.y
59,61c59,61
< int yyerror(const char *errstr);
< void verror(const char *msgP, ...);
< void vwarn(const char *msgP, ...);
---
> void yyerror();
> void verror(char *msgP, ...);
> void vwarn(char *msgP, ...);
906c906
< vwarn(const char *msgP, ...)
---
> vwarn(char *msgP, ...)
921c921
< verror(const char *msgP, ...)
---
> verror(char *msgP, ...)
934,935c934,935
< int
< yyerror(const char *errstr)
---
> void
> yyerror(char *errstr)
940d939
< return 0;
 
plong@mini AM1 % diff lexfuncs.c lexfuncs.orig.c
170c170
< if( (symP = sym_find(&(localStack[i]->symRootP), nameP) ))
---
 
________________________________
From: pid...@googlegroups.com <pid...@googlegroups.com> on behalf of Bill E <wjegr...@gmail.com>
Sent: Tuesday, December 30, 2025 12:18
To: [PiDP-1] <pid...@googlegroups.com>
Subject: Re: [PiDP-1] Re: AM1 update posted to github
 
 
 
On Monday, December 29, 2025 at 6:12:23 PM UTC-5 pl... wrote:
I just built this under MacOS and it needed a few tweaks - so while ymmv here they are:
 
Interesting. Apparently MacOS has some strange ideas about things, at least its C compiler does.
The yyerror decl is consistent in parser.y and matches what Bison says it should be, not sure where the int bit is coming from. Online docs say either is acceptable, though.
The eval.c no return is also bogus. From what I see, the only non-return cases are if yyerror is called, which never returns, it exits. Of course the compiler doesn't know that, but I do.
Is it insisting on const, or is that just a personal preference? I've never bothered with it, too much 'training wheels' aspect for me.
I did add ./ to symtabgen. I never thought about that, I've always (since Unix early days) had . in my path.
 
Bill
--
You received this message because you are subscribed to the Google Groups "[PiDP-1]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-1+un...@googlegroups.com<mailto:pidp-1+unsubscribe@googlegroups.com>.
To view this discussion visit https://groups.google.com/d/msgid/pidp-1/fae68611-a839-406c-b327-a44022cc6fd2n%40googlegroups.com<https://groups.google.com/d/msgid/pidp-1/fae68611-a839-406c-b327-a44022cc6fd2n%40googlegroups.com?utm_medium=email&utm_source=footer>.
sunnyboy010101 <sunnybo...@gmail.com>: Dec 29 10:04AM -0800

I'm still missing the standoff for the last LED - supplied my Mike in
Ontario and coming later this week. Thanks Mike!
 
I installed the electronics in my console case today. Here are a shot of
front and back. Yes, the IC's are missing (I always install them last), and
the ribbon cable is 20cm instead of 10cm because I broke a connector (they
are hard to get on!). And, the Pi-5 is missing as well, waiting on my
bottom m.2 NVME hat.
 
Still, I think the console case looks great. I'll use this for initial
testing and then move things over to the rack once it's verified working
well.
 
-R[image: T1280_pidp1 console with electronics 2025-12-29 01.jpg][image:
T1280_pidp1 console with electronics 2025-12-29 02.jpg]
On Sunday, December 28, 2025 at 3:26:54 PM UTC-8 sunnyboy010101 wrote:
 
sunnyboy010101 <sunnybo...@gmail.com>: Dec 29 08:36PM -0800

My Raspberry Pi NVME bottom hat arrived today, so I moved the SSD from the
top hat to the bottom hat. After some debugging (the NVME would not show up
at all) and a bit of reconfiguring, it fired up and recognized the SSD
which I'd set up as the virtual PiDP1.
 
I plugged it into the connector in the back of my console, and then fired
it up. Of course the wireless failed (it has a bad habit of doing that). I
plugged in a network cable and connected to the Pi-5 no problems. I ran
/opt/pidp1/test.sh and it worked perfectly! (yes, there's a missing LED but
I knew it would not be an issue to the test... it just won't light up). All
installed LEDS worked perfectly, and testing the switches showed them all
to be fully functional. Yay!
 
I then had to read the manual to find out how to get the 'real' PiDP-1
operating instead of the virtual panel which turned out to be a simple
command. I then turned it on, loaded the snowflake tape and hit "load in".
It loaded and the snowflake pattern appeared on my screen on the VNC
connection to the Pi-5.
 
All is working as expected. The console version is a nice, neat little
package that for now sits on my desk in front of the monitor (cabling
issues with the network cable and power). Once I get the LED spacer and LED
installed, I'll move it over to the rack and fire it up.
 
I have some photos and two videos (test.sh & snowflake) that I'll put
somewhere (probably youtube) and link on my web pages as well as here.
 
-R
 
On Monday, December 29, 2025 at 10:04:31 AM UTC-8 sunnyboy010101 wrote:
 
Bill E <wjegr...@gmail.com>: Dec 29 01:07PM -0800

From my Computer History Museum visit. There is a vid of Peter Sampson
loading the music program in preparation for about 40 minutes of Xmas
songs, but can't post it here, too big. It's in the root of my pdp1-mods
repo if you want to see a PDP-1 in action.
 
Bill
Bill E <wjegr...@gmail.com>: Dec 29 02:12PM -0800

I replaced the vid in the repo with a longer mp4 version, much better. It
has audio.
Bill
Glenn Babecki <glenn....@gmail.com>: Dec 29 05:13PM -0500

And for not the real thing...
 
Our daughter was recently up in NYC at the Museum of the Moving Image and
came across an exhibit related to (as she put it) "ancient" video games.
She sent an attached picture of this "PDP-1" exhibit. My first reaction
looking at the picture on the phone was "Wait, What?!". This can't be as
the only PDP-1 exhibit I know is at the CHM.
 
Then upon closer inspection, this exhibit appears to be a PiDP-1 hybrid
thing (i.e., no hulking cabinets). The Flexowriter and the paper tape
reader are what threw me off at first glance. I haven't attempted to
confirm the composition of the items by scaling things in the photo. I
can't read the poster or find the exhibit details on the museum website. I
don't know when the exhibit went online, but I can't believe someone
cobbled together one of Michael Gardi's Type 30 displays to add to the
exhibit. My daughter did say she thought they "had to do some custom work
to make it museum worthy."
 
Glenn
 
 
 
 
Bill E <wjegr...@gmail.com>: Dec 29 09:15AM -0800

Yes, AM1 'punches' a loader before the program, the same as macro and
macro1 do. However, it's a different loader that handles extended memory
loading. It resides at 7751, same as the usual bin loader.
AM1 allows a program to have code in any of the memory banks, controlled by
a 'bank n' directive in the source. The loader will then put the code in
the correct banks.
Bill
Matthias Barthel <maklumatp...@gmail.com>: Dec 29 06:35PM +0100

I'm curious if your rimloader works with my esp32 simulator. In my
simulator, I have installed the readin mode as faithfully as possible.
 
A special readin mode reads the rimloader from the tape and then the
rimloader is started by the CPU and reads the rest of the tape into the
memory and starts the program at the end .
 
Matthias
 
Bill E <wjegr...@gmail.com>: Dec 29 09:47AM -0800

My extended memory bin loader is in rim format, so an AM1 produced binary
loads via readin as usual. When my bin loader starts, it does enable
extended memory via the eem instruction. Otherwise, it is similar to the
existing bin loader, but not compatible with that format since it has to
deal with extended addresses.
 
Here it is. Note no checksum, we're not reading from electromechanical
devices, no need for checksums.
 
 
lai=760040
 
7751/
eem
loop, rpb
lai
and (177777
dac addr
spi
jmp i addr /done
rpb
dio end
load, rpb
dio i addr
idx addr
sas end
jmp load
jmp loop
addr, 0
end, 0
constants
start 7751
Bill
Matthias Barthel <maklumatp...@gmail.com>: Dec 29 06:51PM +0100

Many thanks 👍
 
Is it possible to build only the AM1 crossassembler from your git ?
 
Matthias
 
Bill E <wjegr...@gmail.com>: Dec 29 12:38PM -0800

Yes. Easiest way is to get the entire project via zip, unzip it. Then go to
the Tools/AM1 directory and just type 'make'.
If you want to use the include files I've already defined, they are in the
Am1Includes dir. You will find other useful tools in the Tools dir also.
Documentation is in the Docs directory.
Bill
John Stout <cuspco...@gmail.com>: Dec 29 11:50AM -0800

I just can't get LISP to run from these instructions Oscar.
 
I set EXTEND switch up, set the lisp.rim tape, and press READ IN (the tape
reads in)
I put 7750 in the TW switches, press CONTINUE
I put 400 in the TW switches, press CONTINUE
I set SENSE SWITCH 5 up and ADDRESS to 4, press CONTINUE
 
I can type into the terminal window but typing NIL<space> doesn't do
anything. I've tried with START/CONTINUE but again nothing seems to work.
 
Running test.sh convinces me that all the switches are working.
 
I've tried with both the supplied lisp.rim and that in the archive (Lisp
for the PDP-1 (source) https://simh.trailing-edge.com/kits/lispswre.zip (NB:
this contains a RIM file)) but no joy.
 
I think all the other software seems to work fine (I spent an interesting
hour or so playing Spacewar with my 10-year old grandson yesterday; despite
being an iPad/Super Mario on the Switch child, he loved Spacewar (and
consistently beat me), and just loved being able to do a gravitational
slighshot around the start!): I've assembled and run the circle
demonstration for MACRO, experimented with Minskytron, Pong and Munching
Sqiuares, but LISP doesn't want to play.
 
Do you have any obvious things I can try? The sound isn't working properly
but I'm assuming that's unlikely to be relevant, and proably a soldering
issue.
 
John
John Stout <cuspco...@gmail.com>: Dec 29 11:55AM -0800

Personally I don't see the problem! I think I've 14 or 15 of those, plus
Structure and Interpretation of Computer Programs (which I can't see).
 
I still haven't really got to grips with Let Over Lambda but I keep on
trying!
 
John
 
sunnyboy010101 <sunnybo...@gmail.com>: Dec 29 09:34AM -0800

Most important, the I/O panel DOES NOT use spacers. The LEDs attach
directly to the circuit board, then the panel is fitted to ensure alignment
of the LEDs, then you flip it over (I taped things together first) and
solder one leg (I choose the longer one) of each LED, then flip and check
fitment before soldering the other and clipping off excess leads.
 
I'm surprised you were missing all the standoffs for the main operator
panel (119 offsets for 119 LEDs). Contact Jose using the email in the
instructions to request that many.
 
I did find an amazon.ca (I'm in Canada) link, but you buy 100 or 1000 (two
offers). Both appear to be the correct (8mm) ones with two holes in the
bottom. Here's the amazon link for 1000:
 
https://www.amazon.ca/s?k=led+standoff+8mm&crid=1KWJRIBHKXUYG&sprefix=led+standoff+8mm%2Caps%2C161&ref=nb_sb_noss_1
 
and here's the link for 100:
 
https://www.amazon.ca/Aexit-Spacers-Standoffs-Cylinder-Support/dp/B07L9N4R7R/ref=sr_1_7?crid=1KWJRIBHKXUYG&dib=eyJ2IjoiMSJ9.hn1wQ2ui4C9A9RYOrBEZHiaJye-LTA9GccmkDzN7DabM8yTVXg2qO0mPYYS5N9aILhrPq7g4NzMzq37cD11AMllh7Y1YPdtrIynFNq84a2604tdAhUW9LBUy1bPfsqbFggboKmExbuZ3ZAz8Nn4QpXzvmIcwocwpk-BbD8Ocao1sBsA0B6NnXaQa_JHQqbodLXfHD0Ho1FNYlFNMTPS6Jm9JapAgk1FwGyhEknoUN6tFmvQ11I0UuqrRVg8Vy7_asgBnY7hW52Vr8DukqcThNoFpH5vhPFe1jg9WWXOjikM.cv6f PtJ7WVkBfvHJYAU7bR_Xv6eJZflMrn-IgqbaDXs&dib_tag=se&keywords=led+standoff+8mm&qid=1767029515&sprefix=led+standoff+8mm%2Caps%2C161&sr=8-7
 
I like the photo of the 1000 better than the photo of the 100. Prices are
low and comparable (14.34 CDN & 10 CDN ) but shipping on 100 is more. Both
come mid-to-late January. For 119, I'd personally go with Jose and the
'official' spacers for the kit.
 
-R
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to pidp-1+un...@googlegroups.com.

Bill E

unread,
Dec 30, 2025, 4:50:35 PM12/30/25
to [PiDP-1]
That's the full recording of the snippet I took. This one was done by a recording team that was there. I'm in it in a few places. :)
It was great fun.
Bill

Matthias Barthel

unread,
Dec 30, 2025, 4:53:05 PM12/30/25
to Bill E, [PiDP-1]
Thanks for the Video 🙏 it was great to see the sing-a-long 👍 and the working papertape reader - wonderful!

Matthias 

--
You received this message because you are subscribed to the Google Groups "[PiDP-1]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-1+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-1/a6cfa781-7136-4221-ab75-46d17cd5c055n%40googlegroups.com.

Michael J. Kupec

unread,
Dec 30, 2025, 5:03:22 PM12/30/25
to Michael....@gmail.com, pid...@googlegroups.com
WOW!  Was the PDP-1 cabinet really that deep? Was always under the impression it was the typical 19" rack depth like the PDP-8's were in.  

Are there any pictures of the inside of a typical PDP-1?

Have a great Day!
Michael J Kupec 

To unsubscribe from this group and stop receiving emails from it, send an email to pidp-1+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-1/CAOaX04OqgGVpvHGQtku0o2j_u5qrKKb%3DcDHmGjYuxdL5-u6T8g%40mail.gmail.com.


--
Michael J. Kupec
michae...@gmail.com
http://www.blueovalcorral.com

Bill E

unread,
Dec 30, 2025, 5:37:20 PM12/30/25
to [PiDP-1]
Yep, it's 4 racks. Which was compact for the time. They're full of multiple power supplies and a whole lot of little cards, 'FlipChips' in DEC parlance, and a few miles of wiring.
I got to poke around a bit inside it, fascinating. 

Also amazing, it used less than 2kw. The IBM 1401 from the same general era in a common config used 12kw. The 1401 was running at CHM also, filled a room.
Bill

Glenn Babecki

unread,
Dec 30, 2025, 5:49:58 PM12/30/25
to Bill E, [PiDP-1]
Michael,

I know there are a few video clips from the CHM, but I couldn't find anything extensive. Just search for stuff from CHM, especially with Lyle Blickley.  Sorry, the best I could do without going down the internet rabbit hole.

Bill,

I was having trouble finding the details of your CHM trip, but did you take any pictures in the cabinet?

--
You received this message because you are subscribed to the Google Groups "[PiDP-1]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-1+un...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-1/9050f00a-9ab5-4443-8102-619dcd179043n%40googlegroups.com.

Bradford Miller

unread,
Dec 30, 2025, 6:26:13 PM12/30/25
to Bill E, [PiDP-1]
I think it’s a great example of the “innovators dilemma”… only 95(?!) PDP-1s were sold at the time, and what, 12K 1401s (per Wikipedia). Yet the -1 only needed 36 square feet of space, 110v power, and no special AC or floor reinforcement (or raised floors, presumably), per the -1 installation manual. And cost a small fraction of the 1401.

Now to be fair, the CHM is including a lot more peripherals with the 1401s they have, but it’s still a great case of building something that addresses the bottom of the market, yet grows to overtake those who are addressing the top and trying to preserve their profits. 

Yet despite that, IBM exists and DEC doesn’t today. 

--
You received this message because you are subscribed to the Google Groups "[PiDP-1]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-1+un...@googlegroups.com.

AB

unread,
Jan 2, 2026, 7:10:17 PM (13 days ago) Jan 2
to [PiDP-1]
" Yet despite that, IBM exists and DEC doesn’t today " ... yup, that's the reality
Reply all
Reply to author
Forward
0 new messages