who can use a Harris RTX-2000 processor

[dfn...@gmail.com]

Hi all,

I have an old controller board that contains a Harris RTX-2000 processor. It's a PGA package, so not exactly breadboard friendly, but I thought someone here may want it. The board worked when removed from service in the 90's, so the processor is most likely good.

This is a stack processor. I had a manual for it, but I sent it to someone in the comp.lang.forth group to scan perhaps a decade ago or more. I imagine it's online somewhere.

Would anyone here want it? Priority for someone who may actually be able to use it, unlikely as that may be.

Cheers,

Dave

[menti...@gmail.com]

I don't want it myself, but my Forth AI -- http://dl.acm.org/citation.cfm?doid=307824.307853 -- might want it after sufficient further AI evolution. Any other takers?

ATM
--
http://medium.com/p/12c25b2570b2 -- How Strong AI recognizes an image

[JUERGEN]

Hi Dave,. we want to revive the RTX in FPGA as IP, so I would be very interested - would be to a UK address. It would help to have the real thing when designing the IP for FPGA - probably the Lattice Brevia Board first. When we have the board we would try to get all of the documentation scanned and ready - then partially part of the IP documentation.

[Clive Arthur]

I have a surplus MPE STE Powerboard with RTX2001AJC-8 in plcc if you're
interested, and I'm in the UK, but it wouldn't be free.

I also have about ten RTX2000/2001 chips (some of each - can't remember
the ratio) and a few PTSC1000's.

Cheers
--
Clive

[rickman]

What can you tell us about the board it is on? You call it a
"controller board". Is this a commercial product intended for a
specific application rather than a evaluation board?

I'm a hardware oriented EE, so I have the skills to get it working
although I have never considered a project using this device.

--

Rick C

[Paul Rubin]

rickman <gnu...@gmail.com> writes:
> I'm a hardware oriented EE, so I have the skills to get it working
> although I have never considered a project using this device.

I think the idea was to find a taker for the chip itself, with the
caveat that it might be difficult to remove it from the board. Someone
else mentioned a board with a socketed chip.

I don't think the parts are extreme rarities, and I doubt anyone wants
to build computers out of old units, but I can understand why having
some real ones would be useful to someone trying to develop an FPGA
emulation.

[rickman]

If it is on a board, that is leaps and bounds ahead of having a chip in
your hand. The board has to be set up for doing something and that can
always be of benefit. That's why a hardware background is an advantage.

I can see that Jurgen's friends would have more utility for the unit. I
offered to work with them once but I don't recall there was much I could
do for them. I'm good with VHDL, but I prefer to work from
specifications rather than trying to divine a spec from poor
documentation. I'm not clear on what is available.

--

Rick C

[Paul Rubin]

The RTX-2000 was space qualified if I'm not confusing it with something
else, so I'd expect it to be documented out the wazoo. Maybe it's a
matter of finding some manuals for it and working from those. If it
were me, I'd start with a software emulation. Ymmv.

Why is the RTX-2000 interesting anyway? Do people have old code they
want to run on it? Do they want to write new code for a stack processor
just because it's cool? If the latter, exact compatibility with the
RTX-2000 isn't as important. For that matter, there's always the J1.

[rickman]

On 4/12/2017 12:04 AM, Paul Rubin wrote:
> rickman <gnu...@gmail.com> writes:
>> I'm good with VHDL, but I prefer to work from specifications rather
>> than trying to divine a spec from poor documentation. I'm not clear
>> on what is available.
>
> The RTX-2000 was space qualified if I'm not confusing it with something
> else, so I'd expect it to be documented out the wazoo. Maybe it's a
> matter of finding some manuals for it and working from those. If it
> were me, I'd start with a software emulation. Ymmv.

I'm sure there was lots of documentation. I don't have any of it except
for one document describing the instruction set which I seem to recall
was a bit terse. I don't recall seeing anything describing how it
interfaced with the outside world.

> Why is the RTX-2000 interesting anyway? Do people have old code they
> want to run on it? Do they want to write new code for a stack processor
> just because it's cool? If the latter, exact compatibility with the
> RTX-2000 isn't as important. For that matter, there's always the J1.

The RTX-2000 had a complex machine instruction that could control
separate parts of the CPU to allow multiple data movements in one
instruction. Again, I don't recall the details, but people often refer
to it as supporting multiple instructions in one machine instruction. I
think that is a bit of an exaggeration, but perhaps technically true for
a small set of parallel capable instructions.

--

Rick C

[Lars Brinkhoff]

Paul Rubin wrote:
> Why is the RTX-2000 interesting anyway? Do people have old code they
> want to run on it?

Maybe CMForth?

[lehs]

https://users.ece.cmu.edu/~koopman/stack_computers/sec4_5.html

[Clive Arthur]

On 12/04/2017 05:04, Paul Rubin wrote:
> rickman <gnu...@gmail.com> writes:
>> I'm good with VHDL, but I prefer to work from specifications rather
>> than trying to divine a spec from poor documentation. I'm not clear
>> on what is available.

https://www.intersil.com/content/dam/Intersil/documents/hs-r/hs-rtx2010rh.pdf

> The RTX-2000 was space qualified if I'm not confusing it with something
> else, so I'd expect it to be documented out the wazoo. Maybe it's a
> matter of finding some manuals for it and working from those. If it
> were me, I'd start with a software emulation. Ymmv.

The RTX2010 was the rad-hard version, but pretty similar otherwise.
IIRC it had a multiplier and barrel shifter whereas the RTX2000 had just
the multiplier. I think both had 256 deep stacks.

The RTX2001 was a cheaper version and had neither barrel shifter nor
multiplier, and smaller 64 deep stacks. I used that one a lot.

> Why is the RTX-2000 interesting anyway? Do people have old code they
> want to run on it? Do they want to write new code for a stack processor
> just because it's cool? If the latter, exact compatibility with the
> RTX-2000 isn't as important. For that matter, there's always the J1.

Cheers
--
Clive

[JUERGEN]

Hi Clive, please contact me by email epld...@aol.com, this would be ideal for us - depending on where you live I might come and pick it up.

[Andrew Haley]

JUERGEN <epld...@aol.com> wrote:
>
> Hi Dave,. we want to revive the RTX in FPGA as IP, so I would be
> very interested - would be to a UK address. It would help to have
> the real thing when designing the IP for FPGA - probably the Lattice
> Brevia Board first. When we have the board we would try to get all
> of the documentation scanned and ready - then partially part of the
> IP documentation.

I presume that you have the full RTX reference manual. If not, send
me a message.

Andrew.

[JUERGEN]

Thanks for the offer Andrew. I will see what arrives and come back to you. Thanks for the help everybody. As you have probably seen, Real Time Forth I just published - EP32 next and let's see when RTX fits in.

[JUERGEN]

On Wednesday, April 12, 2017 at 10:18:42 AM UTC+1, Andrew Haley wrote:

Clive, ( Andrew ), just opened my Easter present. Thanks Clive.
MPE RTX 2001 STE Power Board ISS B, Serial Number 1611
MPE RTX-2001 STE PowerBoard description booklet, A5 size, July 1989 rev 1.00, 38 pages,
Harris RTX 2001A Manual August 1989, 38 pages.
Are there any scans around? Would save me doing it again - and your print quality might be better.
Andrew, scans appreciated, if so please send to epld...@aol.com. Any additional support appreciated. I now start as well to collect any additional info I can get hold of.

[Andrew Haley]

JUERGEN <epld...@aol.com> wrote:
> On Wednesday, April 12, 2017 at 10:18:42 AM UTC+1, Andrew Haley wrote:
>> JUERGEN <epld...@aol.com> wrote:
>> >
>> > Hi Dave,. we want to revive the RTX in FPGA as IP, so I would be
>> > very interested - would be to a UK address. It would help to have
>> > the real thing when designing the IP for FPGA - probably the Lattice
>> > Brevia Board first. When we have the board we would try to get all
>> > of the documentation scanned and ready - then partially part of the
>> > IP documentation.
>>
>> I presume that you have the full RTX reference manual. If not, send
>> me a message.
>

> Clive, ( Andrew ), just opened my Easter present. Thanks Clive.
> MPE RTX 2001 STE Power Board ISS B, Serial Number 1611
> MPE RTX-2001 STE PowerBoard description booklet, A5 size, July 1989 rev 1.00, 38 pages,
> Harris RTX 2001A Manual August 1989, 38 pages.
> Are there any scans around? Would save me doing it again - and your
> print quality might be better.
> Andrew, scans appreciated, if so please send to
> epld...@aol.com. Any additional support appreciated. I now start as
> well to collect any additional info I can get hold of.

The programmer's reference manual is about 130 pages, with full
instruction set details.

Andrew.

[opamp]

If still available, I'd love to take a crack at getting it working. The local Harris HAM group would likely get a kick out of it. I work with a number of ex Intersil folks, sadly they were not involved with any RTX projects. Still trying to track down any locals that worked in or around the RTX fab.

[lrei...@cfl.rr.com]

On Monday, May 29, 2017 at 12:00:43 PM UTC-4, opamp wrote:
> If still available, I'd love to take a crack at getting it working. The local Harris HAM group would likely get a kick out of it. I work with a number of ex Intersil folks, sadly they were not involved with any RTX projects. Still trying to track down any locals that worked in or around the RTX fab.

I have a number of manuals in a Black Harris RTX Real Time Express binder with an illustration of a Train that has RTX on it: RTXDS User Manual, RTXDB User Manual, Programmer Manual, RTXDS Technical Reference RTX2000 Data Sheet. It was given to me by a friend who worked on the project who has passed. You are welcome to it if you want it.

[Jeff Chimene]

Do you have any pointers to these manuals. I'd like to implement the RTX2000 (or 1A) for the open-simh[https://github.com/open-simh] project. I'm missing the rtx2000 hardware reference manual. It may have some useful information.

Thanks for any info you can provide!

[Clive Arthur]

This is for the RTX2010, the rad-hard version...

https://github.com/mschuldt/forth-chips/blob/master/harris/rtx2010rh.pdf

--
Cheers
Clive

[Jeff Chimene]

TY, Clive! I think these series of RTX-2k spec sheets are about as good as it's going to get.

[Jan Coombs]

On Wed, 3 Aug 2022 16:15:55 -0700 (PDT)
Jeff Chimene <jchi...@gmail.com> wrote:

> Do you have any pointers to these manuals. I'd like to implement the RTX2000 (or 1A) for the open-simh[https://github.com/open-simh] project. I'm missing the rtx2000 hardware reference manual. It may have some useful information.

JFAR_6_1 is a good article with opcode details, try looking here:
http://soton.mpeforth.com/flag/jfar/vol6.html

Another soft core:
https://mpeforth.com/arena/rtxcore_brief.pdf

I may still have paper copies of the RTX2000 launch data, but no idea
where.

Jan Coombs
--

[Clive Arthur]

On 04/08/2022 00:15, Jeff Chimene wrote:

<snip>

> Do you have any pointers to these manuals. I'd like to implement the RTX2000 (or 1A) for the open-simh[https://github.com/open-simh] project. I'm missing the rtx2000 hardware reference manual. It may have some useful information.
>
> Thanks for any info you can provide!

I have a scan of the Harris RTX2000 Programmer's Reference Manual dated
August 1898 if that's any use.

--
Cheers
Clive

[Kerr-Mudd, John]

That's what I call retro programming!


--
Bah, and indeed Humbug.

[Clive Arthur]

Oops :-)

August 1989

--
Cheers
Clive

[Jeff Chimene]

No thanks, I have access to the pgmr's ref.
>
> --
> Cheers
> Clive

[Jeff Chimene]

I have posts starting at https://systasis.co/prelude/

Posts are in the format of a commonplace book.

[Shiera Delusa]

On Monday, April 10, 2017 at 3:51:33 AM UTC+9, dfn...@gmail.com wrote:
> Hi all,
>
> I have an old controller board that contains a Harris RTX-2000 processor. It's a PGA package, so not exactly breadboard friendly, but I thought someone here may want it. The board worked when removed from service in the 90's, so the processor is most likely good.
>
> This is a stack processor. I had a manual for it, but I sent it to someone in the comp.lang.forth group to scan perhaps a decade ago or more. I imagine it's online somewhere.
>
> Would anyone here want it? Priority for someone who may actually be able to use it, unlikely as that may be.
>
> Cheers,
>
> Dave


Hi Dave,

I know that this post is almost six years old but I was wondering if this is still available?