Any news?

[frot...@gmail.com]

It has been quite some time since I've heard of news regarding the PiDP-10 project.  What's new?  Are there schematics or PCB designs available?

[Andy]

I don't have any info but I am going to assume that Oscar is still maxed out looking after his mum... in the meantime I've been playing about with TOPS-10 with SIMH on my Raspberry so when the time comes I will hit the ground running!

[johntk...@gmail.com]

I heard a whisper on Facebook. Is the project on again?

[Andy]

wait, what? Do share your whisper!! 

cheers

Andy

[johntk...@gmail.com]

Lars posted that Oscar was back to work on it..

[Lars Brinkhoff]

Here is some proof:

https://www.youtube.com/watch?v=VNINZwO_GuA

[Andy]

Very exciting news Lars, Oscar!

Andy

[oscarv]

Thanks! Yes, Lars and I spent a very enjoyable week to get simh to blink the lights properly. I am afraid my brain was not quite up to the job, but now the prototypes (Lars and I both have one now) blink their front panel as we think they should. With that, the software side is pretty close to final (I am sure improvements will come along).

The main thing now is to have the injection mold for the case done. This (together with health problems in the family) is why the project was dormant for the last two years. But that phase is over: I have to get the case produced now quickly. In a somewhat crazy turn of events, the PiDP-10 will be sort of a development project. It will be made in the Panamanian jungle and a few people really depend on me finishing that darned case quickly to get that going. I'll have to elaborate in a longer post I guess :-)

If anyone can recommend

- a good 3D modeller to perfect the case

- anyone with injection molding expertise (I gambled my way through with the PiDP-11 project, but this one is a bit more costly so gambling is not the right approach....)

.... I would be very eager to hear!

Here is the draft case (not faithful enough yet to turn into a mold), next to how it should look RSN:

Kind regards,

Oscar.

[wjegr...@gmail.com]

That is just gorgeous. I'm so glad you did decide to go all-in and do a full replica. Can't wait to order one.

Bill

[a...@papnet.eu]

Hm, i tried to reply before but failed. the user interface of google groups is quite terrible.

Anyway: I also made a KA10 front panel earlier this year, but polygonal. Maybe it will also be useful as a reference: http://aap.papnet.eu/pub/pdp/10/KA10_panel.zip (obj format)

[Antti Louko]

Count me in as well!

I would have a suggestion: Please share the schematics of the panel electronics.  Depending how it is implemented, I might have some ideas to have it support USB interface to a PC to have an option besides the Raspi connection.  I have a running PDP-10 in my server and I would like to have its panel in my work area.

If that is not practical, I may of course build a attachment which fits to the Raspi connector.

[Mike Ross]

Wait, what? This is going to be a full-sized replica?! That's what I've been urging all along! FAR better than a scaled down thing.  Count me in! Put me down for five! Heck, talk to me if you need a bigger minimum order to get the ball rolling!

Mike

[Lars Brinkhoff]

No, it's smaller.  I believe it's the same 6:10 scale as the previous PiDP replicas.

[oscarv]

Angelo,

Thanks! I opened your model in Blender and you've got the svelte curves at the side of the machine somewhat better than I have. So very useful!

I still think I need a professional 3D modeller to sit next to the original bezel to make an exact replica of the case. The injection mold is so expensive that it must be perfect. Incorporate the flaws of the original :)

Still looking how to arrange it all.

Kind regards,

Oscar.

[oscarv]

Antti,

I can send you the Kicad project files if you want. But perhaps wait for two weeks as I just started a revision of the PCB which is in the middle of the not-quite-finished stage at the moment :-)

I think there will be two ways to use the PiDP-10. My software version would contain a Pi that simulates the KA10 (and KI/KL10, why not cheat a bit) and makes it a stand-alone machine. The second version would be in the spirit of Joerg Hoppe's Blinkenbone. Which means the simh simulator runs on your laptop and sends the blinkenlight data over TCP/IP to the front panel unit. Which, then, would still contain a Pi. That is easier than doing things over USB, although that is possible to (OTG USB used as networking device, very simple).

The Pi is so cheap these days that the only reason to use some other circuitry at the back is that, er, Pi's are not all that available at the moment. But that should change.

Kind regards,

Oscar.

[oscarv]

Mike,

Yes, sorry to disappoint, still scale 2:3 roughly. But even then the unit is 53cm wide. The cost of making these things full-sized is much more than two times the scale 2:3. An injection mold that size would cost as much as a small house... maybe a very small house but still.

Kind regards,

Oscar.

[Andy]

Oscar - the case looks awesome. And yep 53cm is still pretty decent size...

cheers

Andy

[Michael Thompson]

>I think there will be two ways to use the PiDP-10. My software version would contain a Pi that simulates the KA10 (and KI/KL10, why not cheat a bit) and makes it a stand-alone machine."

Don't forget the KS10. Even though it only has a few switches on the front, Simh will still run OK.

You could connect your PDP-11 emulator to the PDP-10 emulator to make a somewhat more authentic KL10 boot experience.

--
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-10/9ad5fac7-62e7-4587-9e24-005017a1fc7fn%40googlegroups.com.

--

Michael Thompson

[Bradford Miller]

Sent from my iPhone

On Oct 14, 2022, at 11:27 AM, Michael Thompson <michael.9...@gmail.com> wrote:



>I think there will be two ways to use the PiDP-10. My software version would contain a Pi that simulates the KA10 (and KI/KL10, why not cheat a bit) and makes it a stand-alone machine."

Don't forget the KS10. Even though it only has a few switches on the front, Simh will still run OK.

You could connect your PDP-11 emulator to the PDP-10 emulator to make a somewhat more authentic KL10 boot experience.

Personally, I would love being able to hook up a PiDP-11 to the 10 for this. Especially if the 11 could then handle I/O for the 10 after booting, as it did under ITS and I believe the 20 (remember, only DEC could add 10+11 and get 20 :-).

On Fri, Oct 14, 2022 at 10:26 AM oscarv <vermeul...@gmail.com> wrote:

Antti,

I can send you the Kicad project files if you want. But perhaps wait for two weeks as I just started a revision of the PCB which is in the middle of the not-quite-finished stage at the moment :-)

I think there will be two ways to use the PiDP-10. My software version would contain a Pi that simulates the KA10 (and KI/KL10, why not cheat a bit) and makes it a stand-alone machine. The second version would be in the spirit of Joerg Hoppe's Blinkenbone. Which means the simh simulator runs on your laptop and sends the blinkenlight data over TCP/IP to the front panel unit. Which, then, would still contain a Pi. That is easier than doing things over USB, although that is possible to (OTG USB used as networking device, very simple).

The Pi is so cheap these days that the only reason to use some other circuitry at the back is that, er, Pi's are not all that available at the moment. But that should change.

Kind regards,

Oscar.

--
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-10/9ad5fac7-62e7-4587-9e24-005017a1fc7fn%40googlegroups.com.

--

Michael Thompson

--
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/CAH1BU%3D_nVwUM4UCgbg%2B_qEvH8AUspUkiKjLUe3PBZuSJb6wZqQ%40mail.gmail.com.

[Lars Brinkhoff]

Bradford Miller wrote:

Michael Thompson <michael.9...@gmail.com> wrote:

You could connect your PDP-11 emulator to the PDP-10 emulator to make a somewhat more authentic KL10 boot experience.

You could, but how would that work?  Run RSX20-F and have it poke the KL10 processor?  That would be great, but is it realistically feasible?

Personally, I would love being able to hook up a PiDP-11 to the 10 for this. Especially if the 11 could then handle I/O for the 10 after booting, as it did under ITS and I believe the 20 (remember, only DEC could add 10+11 and get 20 :-).

AI ITS had three: the TV-11 doing the raster graphics terminals, XGP-11 doing printing, and the IO-11 with Chaosnet.  MC ITS (the KL10) had two: the console 11/40, and another version of IO-11 that also handled disks and terminals.

Getting all this to work would be awesome, but the point remains: it would be a lot of work.  Maybe with the PiDP-10 there will be more enthusiasts with free time on their hands?

[Dave]

For my FPGA PDP-10, I was planning on an embedded -11 to help with the USB protocol.  If I have one, no reason not to have more so a TV-11, XGP-11, and IO-11 is a possibility.  Yeah, a lot of work.

Then there's the question of how to connect to the PiDP-10/11 consoles.  I don't have enough I/O pins from the FPGA to do it directly (and I think the PiDP consoles run on 5V and so would need level converters which is a pain) and that would only allow a single console anyway.  I have put a CAN bus interface on my design for other reasons so at first I thought I'd just build a little board with a CANbus I/O expander to drive a console.  Could then daisy-chain as many consoles as I wanted -- a PiDP-10 followed by three or four PiDP-11s, no problem.  Except I think Oscar is using an I2C (?) port expander in the PiDP-10 so I'm not sure how well stacking those up would work.  Still thinking on this issue.

[wjegr...@gmail.com]

Hmm, my emailed response didn't make it. So, by 'all in' I meant with an injection-molded frame. Early discussions seemed to say that would be too expensive. Glad it's happening.

[oscarv]

>> So, by 'all in' I meant with an injection-molded frame. Early discussions seemed to say that would be too expensive. Glad it's happening.

The prototypes live in a custom wooden case. I thought I could do that and avoid the unpleasantries of injection molding. But I never felt happy about it.

So no, if the quote for the mold comes in at a vaguely reasonable level it will be done that way. From the PiDP-11 I learned to trust that it may take a year or two, but in the end I'll recoup the investment in it. If I trust that there will be 1,000 PiDP-10s, then the per-unit expense of the mold is OK. The buffer from making way more PiDP-11s than I thought is used for this. And if not sufficient, I can always go flip hamburgers at McDonalds for a while to make up for having the machine as I wanted it to be at home. :-)

Kind regards,

Oscar.

[Bradford Miller]

OK, I may have to hook up two (2) PiDP-11s then, so hopefully the bus will have enough connectors ;-)

Will MOBYs be sold separately (they had their own front panels ;-).

[Lars Brinkhoff]

The AI lab moby memory is stored here:

https://www.computerhistory.org/collections/catalog/102731715

Here is a closeup of the panel:

https://www.flickr.com/photos/mwichary/2323461568/in/pool-19179520@N00/

The other devices, such as the Systems Concepts disk and terminal controller, are more difficult.

Maybe this will be a trip down memory lane:

http://its.pdp10.se/pdp6-timeline/

[Bill Ezell]

Fabri-tek, wow, that brings back memories. Heh.

On 10/19/2022 2:45 PM, Lars Brinkhoff wrote:

The AI lab moby memory is stored here:

https://www.computerhistory.org/collections/catalog/102731715

Here is a closeup of the panel:

https://www.flickr.com/photos/mwichary/2323461568/in/pool-19179520@N00/

ThbLM-AyqopGZyGF99r2kXoJ3tyew4v-K9ecZa0_VzGQ%40mail.gmail.com.

-- 
Bill Ezell
----------
I happen to know this is the Lupin Express

[John Mann]

Oscar,

On Sat, 15 Oct 2022 at 01:26, oscarv <vermeul...@gmail.com> wrote:

Antti,

...

The Pi is so cheap these days that the only reason to use some other circuitry at the back is that, er, Pi's are not all that available at the moment. But that should change.

Given current Pi availability problems:

Which model Pi's will work driving a PiDP-10 ?

  Raspberry Pi 4B (which RAM sizes?)

  Raspberry Pi Zero 2 W

  Raspberry Pi 400 + 40 pin ribbon cable
      (whole of Pi 400, or just the circuit board)
  older PI's ??

How about

  Raspberry Pi CM4 + carrier board

  e.g. https://www.dfrobot.com/product-2196.html

Or Pi clones

  Banana Pi (various)

  Orange Pi (various)

   ...

Thanks,

    John

 

[oscarv]

Hi,

On Saturday, October 15, 2022 at 5:58:15 PM UTC+2 dab wrote:

Then there's the question of how to connect to the PiDP-10/11 consoles.  I don't have enough I/O pins from the FPGA to do it directly

I think there should be enough pins - the circuit is multiplexed so - I think from memory - 16+8+5 pins is all you need.

 

(and I think the PiDP consoles run on 5V

No, it will run on 3.3V! There's 5V on-bord but 3.3V I/O pins are fine.

So this would be very feasible and very interesting!

Kind regards,

Oscar

[frot...@gmail.com]

Pi clones support would be excellent, particularly given the gouging that's been going on with the genuine Raspberry Pis.

[Lars Brinkhoff]

Should be workable as long as the header pinout is the same.  Or if it's not the same, at least the same number of GPIOs.

[Andy]

Hey Lars any news from Oscar?