First demonstration of the PiDP-10 at VCF SW, Dallas, June 23-25

[oscarv]

Well, whilst waiting for the production parts to come in, I decided to book a ticket to Dallas, TX. There will be a nice corner in their exhibit space for DEC-related things, so why not take advantage to do a World Premiere?

https://www.vcfsw.org/

Lots of work still to come before all the work is done: one more respin of the PCB, ordering all sorts of parts... but especially the software needs some work now. 

Still, a simple Type 30 display will be part of the show, as will two joysticks to play PDP-10 spacewar. The prototype is presentable enough to entertain.

Maybe I get to meet some of you there!

Kind regards,

Oscar.

[Andrew O'Neill]

Wrong side of the pond, I'm afraid. Best of luck!

--
You received this message because you are subscribed to the Google Groups "PiDP-10" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-10+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-10/f7da6ea9-727e-4c84-b39a-aeee6abe98a6n%40googlegroups.com.

[Jason T]

I will make a bee-line for your table as soon they open the doors!
Your table at VCFMW is available, too, if Chicago is in your plans
this September.

j

[Ethan Dicks]

On Mon, Jun 19, 2023 at 3:47 PM oscarv <vermeul...@gmail.com> wrote:
> Well, whilst waiting for the production parts to come in, I decided to book a ticket to Dallas, TX. There will be a nice corner in their exhibit space for DEC-related things, so why not take advantage to do a World Premiere?

Sorry to miss you. I can't make those dates or that distance.

If you do get to VCF Midwest this September, I'll see you there!

Definitely still interested in a PiDP-10!

-ethan

[Richard C]

Hi Oscar,

    Let me know what software needs work. I sent you several emails asking if things are good and have not heard back from you.

Rich

[Lars Brinkhoff]

There are a few glimpses of the PiDP-10 in this photo album:

https://photos.app.goo.gl/aLmSd6FENWiW8pU78

[Mike Ross]

Hope Oscar hasn't forgotten the 'status' lights on the left side are red!



Mike 

On Sun, Jul 2, 2023 at 5:36 PM Lars Brinkhoff <lars.br...@gmail.com> wrote:

There are a few glimpses of the PiDP-10 in this photo album:

https://photos.app.goo.gl/aLmSd6FENWiW8pU78

--

You received this message because you are subscribed to the Google Groups "PiDP-10" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-10+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-10/aee4c6e8-ef8a-4846-bf99-714a0807c315n%40googlegroups.com.

--

http://www.corestore.org
'No greater love hath a man than he lay down his life for his brother.
Not for millions, not for glory, not for fame.
For one person, in the dark, where no one will ever know or see.'

[wjegr...@gmail.com]

So now that the 10 is a reality, what next? VAX is boring. There is a system that most people don't know about, the Burroughs Bx line. They revolutionized computing back in the '60s., way ahead of their time. So many concepts we now take for granted were first developed by Burroughs. OTOH, front panel could be a problem, not very compact:

[Bradford Miller]

I thought this was just the beginning of the PDP-10 front panel series; we still have the MOBY panels, the diagnostics panel (including the rheostat and meter for adjusting core memory voltage), the magic/more magic toggle switch, ...

On Jul 3, 2023, at 2:03 PM, wjegr...@gmail.com <wjegr...@gmail.com> wrote:

So now that the 10 is a reality, what next? VAX is boring. There is a system that most people don't know about, the Burroughs Bx line. They revolutionized computing back in the '60s., way ahead of their time. So many concepts we now take for granted were first developed by Burroughs. OTOH, front panel could be a problem, not very compact:<b.jpg>

On Sunday, July 2, 2023 at 4:03:07 AM UTC-4 tmfd...@gmail.com wrote:

Hope Oscar hasn't forgotten the 'status' lights on the left side are red!



Mike 

On Sun, Jul 2, 2023 at 5:36 PM Lars Brinkhoff <lars.br...@gmail.com> wrote:

There are a few glimpses of the PiDP-10 in this photo album:

https://photos.app.goo.gl/aLmSd6FENWiW8pU78

-- 

You received this message because you are subscribed to the Google Groups "PiDP-10" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-10+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-10/aee4c6e8-ef8a-4846-bf99-714a0807c315n%40googlegroups.com.

-- 

http://www.corestore.org
'No greater love hath a man than he lay down his life for his brother. 
Not for millions, not for glory, not for fame. 
For one person, in the dark, where no one will ever know or see.'

-- 
You received this message because you are subscribed to the Google Groups "PiDP-10" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-10+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-10/f0e91d2e-012a-4703-ae2f-1fc94e013829n%40googlegroups.com.
<b.jpg>

[wjegr...@gmail.com]

Heh, forgot about those little details. The magic toggle switch was pretty amusing. By MOBY, do you mean that Fabritek 256K (36 bit) core monstrosity for the PDP-6, later 10? AI Lab had at least one. Or, did I miss something else?

[Bradford Miller]

Well I had just meant the PDP-10 256k memory frames (they were full racks as I recall with a display on top). I think 256K was called a “Moby” (after Moby Dick?) for a while. I particularly remember the ones on MC, but they all had them I think. The only part I specifically remember from Fabritek was the page-o-meter(?) that was part of the VM stuff on the 6 but my memory is fuzzy.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-10/5aff2d87-38f3-493a-b549-3ae57a688cc0n%40googlegroups.com.

[terry-...@glaver.org]

On Monday, July 3, 2023 at 3:20:59 PM UTC-4 bradford...@gmail.com wrote:

I thought this was just the beginning of the PDP-10 front panel series; we still have the MOBY panels, the diagnostics panel (including the rheostat and meter for adjusting core memory voltage), the magic/more magic toggle switch, ...

Those would all rely on the internal state of the simulation (be it software or hardware) for the details. Same thing with the 370/135 and larger IBM models and most contemporaneous processors. I don't know of any modern logic gate level simulators and the ones 'back then' were to prove the correctness of the logic design and not for speed. And customer machine front panels were there to help technicians repair them. The 370/115 was naked, the /125 had something like 8 lams and even the /138 which inherited the full front panel from the original 135 design had a fully functional front panel equivalent on the console CRT. And there were additional indicators and switches on hidden panels inside the processor cabinet.

If you want something that just had lots of flashing lights like the 1960s BatCave 'computer' that would be trivial (and I nominate the IBM 360 Model 91 8-), but if you want an accurate representation of what's going on you're going to pretty much need gate-level simulation. Remember, just about every light on the PiDP-11 relates to either a Unibus (or MK11) signal or some externally visible register (I/D, memory addressing mode, etc.) Offhand I can only think of the ucode address that has no equivalent in a simulator since it isn't running 11/70 microcode. Occasionally something gets added to make the simulation closer (generally only to make diagnostics happy), but mostly it is an empty void.

[Peter Long]

If it’s blinken-lights people are after - the Burroughs B6700 MDL is awesome - and since it shows registers, could be done via simh etc - but as they weren’t as popular as the PDPs.

Unlike  Oscar’s PDP-10 (which looks awesome), I'm not sure it’s worth the effort to recreate - so many lights !!!

Anyway, if  you want to take a look - check out http://www.retrocomputingtasmania.com/home/projects/burroughs-b6700-mainframe

Cheers

Peter

--
You received this message because you are subscribed to the Google Groups "PiDP-10" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-10+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-10/84f355eb-7536-4956-b0bd-3c323b99b558n%40googlegroups.com.

[Bradford Miller]

For blinkenlights, it would be hard to beat the AN/FSQ-7. Bits of it ended up in a ton of tv shows and movies in the late 50s and 60s.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-10/3A31CE39-A340-4F11-8B93-61E772F1D712%40insys.com.au.

[terry-...@glaver.org]

On Monday, July 3, 2023 at 10:30:45 PM UTC-4 pl...@insys.com.au wrote:

If it’s blinken-lights people are after - the Burroughs B6700 MDL is awesome - and since it shows registers, could be done via simh etc - but as they weren’t as popular as the PDPs.

Unlike  Oscar’s PDP-10 (which looks awesome), I'm not sure it’s worth the effort to recreate - so many lights !!!

Anyway, if  you want to take a look - check out http://www.retrocomputingtasmania.com/home/projects/burroughs-b6700-mainframe

I think the 360/91 has more lamps. The "lamp test" button on these put a serious strain on these as it turned on every controllable lamp (including on freestanding controllers and peripherals) that supported it.

While I never got to do that on a /91, I did have a pair of linked Gandalf Dual PACX IV terminal concentrators (2 full H960 racks) at SPC (picked it up as a donation, traded it in on a pair of fully-loaded DECserver 550s some years later in a carefully crafted "stacked discounts" deal that made the DECservers essentially free). Each card had LEDs for each port (4 LEDs per port, 4 or 8 ports per card). This is 1970s "miniature" LEDs which were power-hungry and somewhat dim. This was a pair of terminal concentrators, linked by one-off (apparently) Multiwire PCBs (individually-placed wires embedded in resin) that had a combined total of 768 terminal ports and 256 host ports. We never used anywhere near all of them, but it looked very impressive and the department wasn't being billed for power, so what the heck). That worked out to 3000+ LEDs just for the terminal side, never mind the host side. And each of the 4 processors in it had a big white "lamp test" button, all linked together. If you stood with your back to it and someone hit lamp test, you could actually feel the heat (as infrared) coming off of it. It got to be a regular thing until the continued stress caused one of the power supplies to fail. I disconnected the lamp test buttons shortly thereafter. Other than looking very impressive, it is utterly unworthy of being simulated (about all it did was say "Host?" if you pressed Break, counted to 3 and then hit Return on a terminal (Break wakes up the scanner to listen to your port, Return lets it guess your baud rate). Oddly enough, Sandia National Labs did a performance simulator (but not an emulation) for it: https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/DE82013386.xhtml

[Lars Brinkhoff]

There's *the* Moby, which indeed is the Fabritek 256K.  Supposedly the largest memory in the world when delivered in 1966.  By random chance, a former Fabritek engineer saw my posts on Twitter and told the story of when he presented at a conference and met Marvin Minsky, in slippers, who placed an order for the huge memory on a napkin.  It took a year or so to pass acceptance tests.  Photo attached below.

A few years later, core memories of 256K became commonplace.  The term Moby was adopted for any 256K memory.  E.g. most ITS machines had a memory amount of "two mobies".

Regarding the "page-o-meter", perhaps what you remember is the memory management paging device, which had a "map-o-meter".  This is the panel for MIT-ML's pager, with the map-o-meter to the right.

The Fabritek panel.  It's in storage at the Computer History Museum.

[Peter Long]

The one at Monash University had the MCP patch that showed the current time when idle - it just depended on what you put in the relevant registers ! 

Peter

 

--- Original message ---
Subject: Re: [pidp-10] First demonstration of the PiDP-10 at VCF SW, Dallas,June 23-25
From: jim stephens <james.w....@gmail.com>
To: Peter Long <pl...@insys.com.au>
Date: Tuesday, 04/07/2023 4:18 PM

The 6700 idle had a Burroughs B that appeared. Saw a quad processor system with 4 of them in Nashville at a site that converted to Multics. 

On Mon, Jul 3, 2023, 9:30 PM 'Peter Long' via PiDP-10 <pid...@googlegroups.com> wrote:

If it’s blinken-lights people are after - the Burroughs B6700 MDL is awesome - and since it shows registers, could be done via simh etc - but as they weren’t as popular as the PDPs.

Unlike  Oscar’s PDP-10 (which looks awesome), I'm not sure it’s worth the effort to recreate - so many lights !!!

Anyway, if  you want to take a look - check out http://www.retrocomputingtasmania.com/home/projects/burroughs-b6700-mainframe

Cheers

Peter

On 4 Jul 2023, at 12:03 pm, terry-...@glaver.org <terry-...@glaver.org> wrote:

On Monday, July 3, 2023 at 3:20:59 PM UTC-4 bradford...@gmail.com wrote:

I thought this was just the beginning of the PDP-10 front panel series; we still have the MOBY panels, the diagnostics panel (including the rheostat and meter for adjusting core memory voltage), the magic/more magic toggle switch, ...

Those would all rely on the internal state of the simulation (be it software or hardware) for the details. Same thing with the 370/135 and larger IBM models and most contemporaneous processors. I don't know of any modern logic gate level simulators and the ones 'back then' were to prove the correctness of the logic design and not for speed. And customer machine front panels were there to help technicians repair them. The 370/115 was naked, the /125 had something like 8 lams and even the /138 which inherited the full front panel from the original 135 design had a fully functional front panel equivalent on the console CRT. And there were additional indicators and switches on hidden panels inside the processor cabinet.

If you want something that just had lots of flashing lights like the 1960s BatCave 'computer' that would be trivial (and I nominate the IBM 360 Model 91 8-), but if you want an accurate representation of what's going on you're going to pretty much need gate-level simulation. Remember, just about every light on the PiDP-11 relates to either a Unibus (or MK11) signal or some externally visible register (I/D, memory addressing mode, etc.) Offhand I can only think of the ucode address that has no equivalent in a simulator since it isn't running 11/70 microcode. Occasionally something gets added to make the simulation closer (generally only to make diagnostics happy), but mostly it is an empty void.

--
You received this message because you are subscribed to the Google Groups "PiDP-10" group.

To unsubscribe from this group and stop receiving emails from it, send an email to pidp-10+unsubscri...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-10/84f355eb-7536-4956-b0bd-3c323b99b558n%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "PiDP-10" group.

To unsubscribe from this group and stop receiving emails from it, send an email to pidp-10+unsubscri...@googlegroups.com.

[Bradford Miller]

Ah yes, that’s the one. My own memory needs a lamp test, apparently.

On Jul 4, 2023, at 1:37 AM, Lars Brinkhoff <lars.br...@gmail.com> wrote:

There's *the* Moby, which indeed is the Fabritek 256K.  Supposedly the largest memory in the world when delivered in 1966.  By random chance, a former Fabritek engineer saw my posts on Twitter and told the story of when he presented at a conference and met Marvin Minsky, in slippers, who placed an order for the huge memory on a napkin.  It took a year or so to pass acceptance tests.  Photo attached below.

A few years later, core memories of 256K became commonplace.  The term Moby was adopted for any 256K memory.  E.g. most ITS machines had a memory amount of "two mobies".

Regarding the "page-o-meter", perhaps what you remember is the memory management paging device, which had a "map-o-meter".  This is the panel for MIT-ML's pager, with the map-o-meter to the right.

<mit-ml-pager.jpg>

The Fabritek panel.  It's in storage at the Computer History Museum.

<fabri-tek.jpg>

tisdag 4 juli 2023 kl. 01:13:25 UTC+2 skrev bradford...@gmail.com:

Well I had just meant the PDP-10 256k memory frames (they were full racks as I recall with a display on top). I think 256K was called a “Moby” (after Moby Dick?) for a while. I particularly remember the ones on MC, but they all had them I think. The only part I specifically remember from Fabritek was the page-o-meter(?) that was part of the VM stuff on the 6 but my memory is fuzzy.

On Jul 3, 2023, at 4:03 PM, wjegr...@gmail.com <wjegr...@gmail.com> wrote:

Heh, forgot about those little details. The magic toggle switch was pretty amusing. By MOBY, do you mean that Fabritek 256K (36 bit) core monstrosity for the PDP-6, later 10? AI Lab had at least one. Or, did I miss something else?

On Monday, July 3, 2023 at 3:20:59 PM UTC-4 bradford...@gmail.com wrote:

I thought this was just the beginning of the PDP-10 front panel series; we still have the MOBY panels, the diagnostics panel (including the rheostat and meter for adjusting core memory voltage), the magic/more magic toggle switch, ...

--
You received this message because you are subscribed to the Google Groups "PiDP-10" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-10+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-10/cd6f911f-38b4-446f-a92e-4c6f07eaa410n%40googlegroups.com.
<mit-ml-pager.jpg><fabri-tek.jpg>

[Ethan Dicks]

On Tue, Jul 4, 2023 at 1:37 AM Lars Brinkhoff <lars.br...@gmail.com> wrote:
> There's *the* Moby, which indeed is the Fabritek 256K. Supposedly the largest memory in the world when delivered in 1966. By random chance, a former Fabritek engineer saw my posts on Twitter and told the story of when he presented at a conference and met Marvin Minsky, in slippers, who placed an order for the huge memory on a napkin. It took a year or so to pass acceptance tests.

I went digging for some contemporary references. Found Scientific
American September 1966 cover topic "logic", with articles by Minsky
and David Evans and Ivan Sutherland. It's chock full of great stuff,
like an ad for the "PDP-8/S, the $10,000 computer... ...and the new,
big PDP-9" (pp 56-57).

In Evans' article, on page 82, above a photograph of a 16 bit Fabritek
core plane is a chart listing memory prices with core at "10^-1
dollars per bit" (IC memory is $10/bit).

Mighty expensive Moby!

-ethan