• Please review our updated Terms and Rules here

About re-producing PDP-11/70

alphaaxp said:
By the way, how much does a complete PDP-11/70 cost now without I/O included?
Click to expand...
IMO that mostly depends on being in the right place at the right time. Local Pickup Only listings for Unibus PDP-11 systems run fairly inexpensively because shipping such systems is a pain for both the seller and the buyer so there's motivation for both parties to perform a "rescue" and that significantly lowers the selling price. The conventional PDP-11/70 system housed the memory in a second rack, so that increases the problems for everyone. I haven't seen a PDP-11/70 of any sort listed on eBay (even a decdatasystem 570) for quite some time. But essentially all of the large Unibus-based CPUs/systems have shown up on eBay in recent years. So there is turnover. And there are other forums and word-of-mouth among hobbyists.

Where are you located? I infer someplace east of Greenwich, England. Being outside of North America greatly decreases the likelihood of one "falling into your lap".
 
Actually, I am in Beijing, China, where there are hardly any PDP-11s. The seller of M8123 agrees to a $100 make offer, but I still need to communicate with him to have him ship to China😂
I also feel that the original rack was too large and inconvenient. Now I have come up with a new one, which looks a bit similar to the original DEC's 30 inch and 40 inch models
屏幕截图 2024-07-24 192314.png
If using an I/O emulator like Unibone, it can be completely installed, it depends on whether the heat dissipation is sufficient.
As a table or cabinet like this, it looks more coordinated
屏幕截图 2024-07-24 193617.png
 
pbirkel@gmail.com said:
I was simply pointing out that the series of "PDP-11" models includes the PDP-11/780. The native ISP is different, and although there is a compatibility mode that doesn't allow one to run, say, RSX-11. ...
Click to expand...

There really is no such thing as a PDP-11/780. That terminology was never used. It always has been VAX-11/780 ...
 
pbirkel@gmail.com said:
I'm sure that I've seen it identified that way in early marketing literature. I'll look around for an example.
Click to expand...
It is mentioned in several documents, for example:

"Additionally, the VAX-11/780 is priced nearly the same as a PDP-11/70 and executes most programs somewhat faster than a PDP-11/780."

See also:
or:
 
Last edited:
jplr said:
It is mentioned in several documents, for example:

"Additionally, the VAX-11/780 is priced nearly the same as a PDP-11/70 and executes most programs somewhat faster than a PDP-11/780."
Click to expand...
The above reference is just a simple typo that occurs once in the abstract of the document. PDP-11/780 should have been PDP-11/70, otherwise the sentence makes no sense.
jplr said:
See also:
Click to expand...
I did not find any PDP-11/780 reference in the above document. Tons of references to all other kinds of VAX models 85xx/86xx/87xx/88xx tho.
jplr said:
or:
Click to expand...
There are some references to PDP-11/780 in the above document but it was written by third parties that obviously did not know what they were talking about. Some MIL contractors.

DEC never referred to any of the VAX series processors by the PDP-11/7xx or PDP 8xxx etc nomenclature. Always was VAX-11/725,730,750,780,782,785 or just VAX 8xxx for the later models.
If you google for "PDP-11/780" yeah there are references by random people using that terminology but they generally are ill informed randos.

I expected you to cite official DEC documents that referred to a PDP-11/780 but I never heard the VAX referred to by that nomenclature back in the day when I was at DEC from 1975-82.
 
pbirkel@gmail.com said:
I was simply pointing out that the series of "PDP-11" models includes the PDP-11/780. The native ISP is different, and although there is a compatibility mode that doesn't allow one to run, say, RSX-11. Again it's important to define your design constraints; so another seems to be to be able to natively run one or more of RSX-11, RSTS, etc.

By another interpretation of "large", the PDP-11/44 would be the correct choice since it replaced the PDP-11/70 and has roughly twice the performance (depending on your metrics of choice) ... and a *much* easier to implement memory subsystem.

I suspect that you have a fourth design constraint: that the model must include a traditional lights-and-switches front panel but I don't think that you've mentioned that anywhere.

IMO there are a limited number of hobbyists ready to mount and sustain an 18U 19" racked CPU. If your objectives include "provide everyone with a real old style minicomputer experience in the future" then IMO that's going to be a very small set of "everyone". I'd consider elaborating on what you define to be a "real old style minicomputer experience" given that it sounds like you intend to cut corners on I/O equipment (e.g., RM03) and memory (MJ11 or MK11).

I'm not criticizing; just trying to draw out and clarify your objectives and constraints.
Click to expand...

The 11/44 is not twice the performance of the 11/70; in fact the 11/70 is in the range of 1.5X to 2X the performance of the 11/44.
The 1981 PDP-11 Processor Handbook on bitsavers has in appendix B detailed instruction timing calculations for the 11/44 and 11/70.
A basic MOV Rx,Ry on the 11/70 is two microcycles of 150ns or 300ns. On the 11/44 the same operation is 600ns.
More advanced address modes are a bit more convoluted but the 11/70 has the faster microcycle timing advantage so it wins.
 
Hunta said:
Page 5, right column
Click to expand...
Ok, I see that. Again, not a DEC document, written by some rando at DataPro Research Corporation back in 1988 where they obviously did not know the correct terminology to use.
By 1988 for certain there was no such system as a PDP-11/780 and there never was starting back into 1978. DEC never called the VAX series processors by PDP nomenclature.
All the other processors PDP-1 thru PDP-15 were all PDPs. The VAX was never a PDP, sorry.
 
> I expected you to cite official DEC documents that referred to a PDP-11/780 but I never heard the VAX referred to by that nomenclature back in the day when I was at DEC from 1975-82.

Were there no "special models" for niche markets at DEC, Paul seems to hint at that? Or even a bit of secrecy/waring sometimes between teams?

Hampage.hu says about the VAX "In the first months of the project, the machine was called PDP-11/780."
Indeed the person writing Hampage.hu was born in 1975 (if I understand correctly) but they seem well informed.
 
jplr said:
> I expected you to cite official DEC documents that referred to a PDP-11/780 but I never heard the VAX referred to by that nomenclature back in the day when I was at DEC from 1975-82.

Were there no "special models" for niche markets at DEC, Paul seems to hint at that? Or even a bit of secrecy/waring sometimes between teams?

Hampage.hu says about the VAX "In the first months of the project, the machine was called PDP-11/780."
Indeed the person writing Hampage.hu was born in 1975 (if I understand correctly) but they seem well informed.
Click to expand...
I think this is just fanciful reverse engineering of history. His page cites no documents, it is just 'facts' with lots of use of the word 'probably'.
When the VAX program started there was no model number, it was just referred to as the STAR hardware prototype in the VAX program.
Not until well along in the development did marketing get involved and come up with the VAX-11/780 moniker. The '78' portion pays homage to the release date in 1978.
 
AK6DN said:
I think this is just fanciful reverse engineering of history. His page cites no documents, it is just 'facts' with lots of use of the word 'probably'.
When the VAX program started there was no model number, it was just referred to as the STAR hardware prototype in the VAX program.
Not until well along in the development did marketing get involved and come up with the VAX-11/780 moniker. The '78' portion pays homage to the release date in 1978.
Click to expand...
I'm coming to the same conclusion. That "first couple of months" claim shows up in several places but no citations to primary sources. Possibly there's a CHM interview with a member of the design team that would substantiate (or refute) the claim. I can certainly imagine an internal codename in advance of marketing getting involved, and "STAR" seems to have been it.

It remains interesting that "Computer Engineering: A DEC View of Hardware Systems Design" (Bell, Mudge & McNamara) repeatedly and consistently uses statements such as "The latest member of the PDP-11 family, the Virtual Address Extension 11 or VAX-11". The emphasis is repeatedly on continuity within a single PDP-11 "family".

I always found it interesting that despite the fact that there was from the get-go a plan for a family of PDP-11 systems, the first production front panels were simply "PDP-11" (no model identification) and only later did the model designation "PDP-11/20" get applied.
 
AK6DN said:
The 11/44 is not twice the performance of the 11/70; in fact the 11/70 is in the range of 1.5X to 2X the performance of the 11/44.
The 1981 PDP-11 Processor Handbook on bitsavers has in appendix B detailed instruction timing calculations for the 11/44 and 11/70.
A basic MOV Rx,Ry on the 11/70 is two microcycles of 150ns or 300ns. On the 11/44 the same operation is 600ns.
More advanced address modes are a bit more convoluted but the 11/70 has the faster microcycle timing advantage so it wins.
Click to expand...
Thank you for the correction and the specific CPU performance reference!


Of course the specialized memory and Massbus interfaces with the 11/70 add a lot of "large system" capability that the 11/44 lacks.
 
pbirkel@gmail.com said:
I can certainly imagine an internal codename in advance of marketing getting involved, and "STAR" seems to have been it.
Click to expand...
I have a recollection of an interview that indicated the original codename was VAX, but when marketing got involved they felt that was giving away too much information and settled on STAR instead. This is borne out by the architecture teams being named "VAX-A", "VAX-B" and "VAX-C". Unfortunately, I haven't been able to find where I saw that, though it was likely in the Supnik or Lary oral histories.

I wish CHM had more stuff published on the web rather than "in the collection", ie unavailable for the un-washed masses to peruse.

CRW

Side note: It would be nice to split this off from the PDP-11/70 clone thread.
 
Bitly said:
I have a recollection of an interview that indicated the original codename was VAX, but when marketing got involved they felt that was giving away too much information and settled on STAR instead. This is borne out by the architecture teams being named "VAX-A", "VAX-B" and "VAX-C". Unfortunately, I haven't been able to find where I saw that, though it was likely in the Supnik or Lary oral histories.
Click to expand...
Which survives in places like STARLET.OLB, at least as far as Itanium VMS. I'm unsure about x86, as (apparently) VSI and I are not on speaking terms. The original product names were:

VAX-11/780 - Star
VAX-11/785 - Supertar
VAX-11/750 - Comet
VAX-11/730 - Nebula
VAX-11/725 - LCN (Low Cost Nebula)
VAX-11/790 (8600) - Venus
VAX-11/795 (8650) - Morningstar

I sense a theme here...

Bitly said:
I wish CHM had more stuff published on the web rather than "in the collection", ie unavailable for the un-washed masses to peruse.
Click to expand...
The charitable view is that they have limited resources which they mainly devote to the exhibits and cataloging donations. The uncharitable view is... unprintable.
 
Terri Kennedy said:
The charitable view is that they have limited resources which they mainly devote to the exhibits and cataloging donations. The uncharitable view is... unprintable.
Click to expand...

There have been a number of people making donations to the document scanning fund, especially since CGB's passing.
Expect more of the DEC corporate archive to become available in the next few years.
 
These days, I have been organizing the chips and substitution table used in PDP-11/70.
The principle is to replace the OC output with the tri state output chip, and replace H, LS, S series chips with F, AS, S series chips

Now I have a question. I have found that some buses are OC outputs, and replacing them may cause compatibility issues.
So, which bus driver chips do DEC generally use, and do they have higher performance replacement components?

What I have found so far is DS8640, and DEC8640 is marked in the engineering drawing.
I have searched for many manuals, but I have not seen one with the same function as it

屏幕截图 2024-07-29 022633.png
 
8640 are the bus transceiver chips for the unibus. They have very specific parameters in order to drive a maximum length unibus.
 
I should also note that while digging around in the basement of my shop I’ve located what appears to be a complete set of 11/70 CPU boards. I also found a set of FP11 (floating point) boards and at least one set of RH70 boards.

Let me know if there’s interest in them and we can work something out.
 
You must log in or register to reply here.
Back
Top