Does anyone in North America or Hawaii still have a Connection Machine
CM-2a that they would like to part with?
The goal if possible, is to put together a working system with a
Symbolics machine. I'm also looking for the host software, interface
board and associated cables.
Thanks,
--Bruce
Robert Bruce Carleton <···@hakuhale.net> wrote in message news:> Does anyone in North America or Hawaii still have a Connection Machine
> CM-2a that they would like to part with?
>
> The goal if possible, is to put together a working system with a
> Symbolics machine. I'm also looking for the host software, interface
> board and associated cables.
Hello Bruce,
I had the fun chance a long time ago to program a CM-1 (a SIMD
machine) using Star-Lisp - really great!
I must say, however, that I would not want to pay the electric bills
for running one in my home.
You might just want to run a dual G5 Mac with LispWorks; lots cheaper
and probably faster :-)
Best regards,
Mark
Hi Mark,
Thanks for the heads up on the size and power requirements of the
Connection Machines. I don't know if you have seen a CM-2a. It looks
like a single cube cut off a full sized CM-2s eight cubes. It's a lot
smaller than a full sized CM-2. There's a picture of one up on the web at:
http://home.hakuhale.net/rbc/symbolics/cm2a-2sm.jpg
I don't figure a connection machine is going to be faster than modern
hardware. I'm just into old lisp hardware I guess.
Best regards,
--Bruce
Mark Watson wrote:
> Robert Bruce Carleton <···@hakuhale.net> wrote in message news:> Does anyone in North America or Hawaii still have a Connection Machine
>
>>CM-2a that they would like to part with?
>>
>>The goal if possible, is to put together a working system with a
>>Symbolics machine. I'm also looking for the host software, interface
>>board and associated cables.
>
>
> Hello Bruce,
>
> I had the fun chance a long time ago to program a CM-1 (a SIMD
> machine) using Star-Lisp - really great!
>
> I must say, however, that I would not want to pay the electric bills
> for running one in my home.
>
> You might just want to run a dual G5 Mac with LispWorks; lots cheaper
> and probably faster :-)
>
> Best regards,
> Mark
On Fri, 27 Aug 2004 17:31:13 -1000, Robert Bruce Carleton
<···@hakuhale.net> wrote:
>Thanks for the heads up on the size and power requirements of the
>Connection Machines. I don't know if you have seen a CM-2a. It looks
>like a single cube cut off a full sized CM-2s eight cubes. It's a lot
>smaller than a full sized CM-2. There's a picture of one up on the web at:
>
> http://home.hakuhale.net/rbc/symbolics/cm2a-2sm.jpg
I played with a small CM-2 in college: 8K processors. The picture
looks right - approximately a 1 meter cube. I remember it throwing
off a lot of heat ... it was kept apart from the other servers in a
special room with its own AC. I definitely wouldn't want the electric
bill!
>I don't figure a connection machine is going to be faster than modern
>hardware.
Depends on the problem, your solution and whether you have a DataVault
8-) There are a number of serial quadratic algorithms that can be
made linear on the CM by using the network to shuffle data. And if
your application really is amenable to SIMD, you'll be amazed at how
fast an 8Mhz machine can execute it when there are thousands of
processors to throw at it.
>I'm just into old lisp hardware I guess.
Unfortunately I didn't get the chance to use *Lisp ... all my programs
were in C*. When university purchased the CM, the latest edition of
*Lisp was in beta and they wanted to wait for it. *Lisp finally
arrived near the end of the semester but it wasn't installed in time
for me to play with it. It was quite difficult to arrange time on the
CM, so after the course in parallel computing ended I never went back
to it.
George
--
for email reply remove "/" from address
I did find a paper about a hybrid CM-2/Xilinx prototype using FPGAs.
Anyone know how much of the SIMD work was done by the FPGA part of the
hybrid?
--Bruce
George Neuner wrote:
> On Fri, 27 Aug 2004 17:31:13 -1000, Robert Bruce Carleton
> <···@hakuhale.net> wrote:
>
>
>>Thanks for the heads up on the size and power requirements of the
>>Connection Machines. I don't know if you have seen a CM-2a. It looks
>>like a single cube cut off a full sized CM-2s eight cubes. It's a lot
>>smaller than a full sized CM-2. There's a picture of one up on the web at:
>>
>> http://home.hakuhale.net/rbc/symbolics/cm2a-2sm.jpg
>
>
> I played with a small CM-2 in college: 8K processors. The picture
> looks right - approximately a 1 meter cube. I remember it throwing
> off a lot of heat ... it was kept apart from the other servers in a
> special room with its own AC. I definitely wouldn't want the electric
> bill!
>
>
>
>>I don't figure a connection machine is going to be faster than modern
>>hardware.
>
>
> Depends on the problem, your solution and whether you have a DataVault
> 8-) There are a number of serial quadratic algorithms that can be
> made linear on the CM by using the network to shuffle data. And if
> your application really is amenable to SIMD, you'll be amazed at how
> fast an 8Mhz machine can execute it when there are thousands of
> processors to throw at it.
>
>
>
>>I'm just into old lisp hardware I guess.
>
>
> Unfortunately I didn't get the chance to use *Lisp ... all my programs
> were in C*. When university purchased the CM, the latest edition of
> *Lisp was in beta and they wanted to wait for it. *Lisp finally
> arrived near the end of the semester but it wasn't installed in time
> for me to play with it. It was quite difficult to arrange time on the
> CM, so after the course in parallel computing ended I never went back
> to it.
>
> George
I have page of Patents with description of CM-2 Chip CPU and linkway switch
system.
Massimo Spataro
"Robert Bruce Carleton" <···@hakuhale.net> ha scritto nel messaggio
····················@corp.supernews.com...
> I did find a paper about a hybrid CM-2/Xilinx prototype using FPGAs.
> Anyone know how much of the SIMD work was done by the FPGA part of the
> hybrid?
>
> --Bruce
>
>
> George Neuner wrote:
> > On Fri, 27 Aug 2004 17:31:13 -1000, Robert Bruce Carleton
> > <···@hakuhale.net> wrote:
> >
> >
> >>Thanks for the heads up on the size and power requirements of the
> >>Connection Machines. I don't know if you have seen a CM-2a. It looks
> >>like a single cube cut off a full sized CM-2s eight cubes. It's a lot
> >>smaller than a full sized CM-2. There's a picture of one up on the web
at:
> >>
> >> http://home.hakuhale.net/rbc/symbolics/cm2a-2sm.jpg
> >
> >
> > I played with a small CM-2 in college: 8K processors. The picture
> > looks right - approximately a 1 meter cube. I remember it throwing
> > off a lot of heat ... it was kept apart from the other servers in a
> > special room with its own AC. I definitely wouldn't want the electric
> > bill!
> >
> >
> >
> >>I don't figure a connection machine is going to be faster than modern
> >>hardware.
> >
> >
> > Depends on the problem, your solution and whether you have a DataVault
> > 8-) There are a number of serial quadratic algorithms that can be
> > made linear on the CM by using the network to shuffle data. And if
> > your application really is amenable to SIMD, you'll be amazed at how
> > fast an 8Mhz machine can execute it when there are thousands of
> > processors to throw at it.
> >
> >
> >
> >>I'm just into old lisp hardware I guess.
> >
> >
> > Unfortunately I didn't get the chance to use *Lisp ... all my programs
> > were in C*. When university purchased the CM, the latest edition of
> > *Lisp was in beta and they wanted to wait for it. *Lisp finally
> > arrived near the end of the semester but it wasn't installed in time
> > for me to play with it. It was quite difficult to arrange time on the
> > CM, so after the course in parallel computing ended I never went back
> > to it.
> >
> > George
George Neuner <·········@comcast.net> writes:
> I played with a small CM-2 in college: 8K processors. The picture
> looks right - approximately a 1 meter cube. I remember it throwing
Was the look of the computer in the movie "War Games" inspired to the
Connection Machine?
Paolo
--
Why Lisp? http://alu.cliki.net/RtL%20Highlight%20Film
Recommended Common Lisp libraries/tools (Google for info on each):
- ASDF/ASDF-INSTALL: system building/installation
- CL-PPCRE: regular expressions
- UFFI: Foreign Function Interface
From: Christopher C. Stacy
Subject: Re: Looking to buy a Connection Machine CM-2a
Date:
Message-ID: <u656yj6rf.fsf@news.dtpq.com>
>>>>> On Wed, 01 Sep 2004 15:57:48 +0200, Paolo Amoroso ("Paolo") writes:
Paolo> George Neuner <·········@comcast.net> writes:
>> I played with a small CM-2 in college: 8K processors. The picture
>> looks right - approximately a 1 meter cube. I remember it throwing
Paolo> Was the look of the computer in the movie "War Games" inspired
Paolo> to the Connection Machine?
I still have a letter (somewhere) from (Jane Fonda's, I think) movie
company that produced WAR GAMES when they wanted to visit the AI lab;
it's been so long that I don't remember if they actually came by or not.
But it was before there was any Connection Machine, so the answer to
your question would be: "No".
See however Jurassic Park.
······@news.dtpq.com (Christopher C. Stacy) writes:
>>>>>> On Wed, 01 Sep 2004 15:57:48 +0200, Paolo Amoroso ("Paolo") writes:
> [...]
> Paolo> Was the look of the computer in the movie "War Games" inspired
> Paolo> to the Connection Machine?
>
> I still have a letter (somewhere) from (Jane Fonda's, I think) movie
> company that produced WAR GAMES when they wanted to visit the AI lab;
> it's been so long that I don't remember if they actually came by or not.
> But it was before there was any Connection Machine, so the answer to
> your question would be: "No".
FWIW, I played that part of _War Games_ with the commentary on (by
the two screenwriters and the director) where they talk about the
computer, but they didn't say anything about how they came up with
its looks. They tell, though, that its real-world prototype was
called SIOP (Single Integrated Operating Procedure), scenarios that
were run by NORAD's supercomputer, about which they read in an
article entitled "The Underground World of the Bomb" by a journalist
called Rosenbaum (if I got the name right). They didn't like that
acronym---wanted something that was "fun and easy to remember"---and
came up with WOPR (War Operation Planned Response), pronounced
"wapper". (_War Games_ was a Leonard Goldberg production in
association with Sherwood Productions.)
By the way, I wonder if any one would by any chance remember the
title of another movie, which must have been made in the late 70s or
early 80s, in which the story was similar in the sense that a
teenager connected (again by chance, I think) to an AI program.
Unfortunately, I only remember two details from it: the acoustic
modems and the program's greeting of "Long time no see"...
---Vassil.
--
Vassil Nikolov <········@poboxes.com>
Hollerith's Law of Docstrings: Everything can be summarized in 72 bytes.
Naw.. Looks more like a suped up Cray Supercoputer to me.
On Wed, 01 Sep 2004 15:57:48 +0200, Paolo Amoroso <·······@mclink.it>
wrote:
> George Neuner <·········@comcast.net> writes:
>
>> I played with a small CM-2 in college: 8K processors. The picture
>> looks right - approximately a 1 meter cube. I remember it throwing
>
> Was the look of the computer in the movie "War Games" inspired to the
> Connection Machine?
>
>
> Paolo
--
Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/
In article <···············@corp.supernews.com>,
Robert Bruce Carleton <···@hakuhale.net> wrote:
> Does anyone in North America or Hawaii still have a Connection Machine
> CM-2a that they would like to part with?
Skef Wholey implemented a version of Connection Machine Lisp
on top of Common Lisp so that you could play with this exotic
dialect on conventional hardware. Check out:
Wholey, S. (1991). An experimental implementation of
connection machine lisp. In P. Lee (Ed.), Topics in
advanced language implementation. Cambridge, MA: MIT
Press.
The code was available at the CMU repository last I
checked.
Sashank,
Thanks for the pointer. I have not seen this paper before.
Best regards,
--Bruce
Sashank Varma wrote:
> In article <···············@corp.supernews.com>,
> Robert Bruce Carleton <···@hakuhale.net> wrote:
>
>
>>Does anyone in North America or Hawaii still have a Connection Machine
>>CM-2a that they would like to part with?
>
>
> Skef Wholey implemented a version of Connection Machine Lisp
> on top of Common Lisp so that you could play with this exotic
> dialect on conventional hardware. Check out:
>
> Wholey, S. (1991). An experimental implementation of
> connection machine lisp. In P. Lee (Ed.), Topics in
> advanced language implementation. Cambridge, MA: MIT
> Press.
>
> The code was available at the CMU repository last I
> checked.
Sashank Varma wrote:
> The code was available at the CMU repository last I checked.
If you're talking about the *Lisp simulator, J. P. Massar has
maintained that code, and a year or so ago released an updated
version --- f20.
--
Fred Gilham ······@csl.sri.com
I think it's pretty obvious that the worship of that false idol known
as the State has, in the 20th Century, had some very bad effects. The
historians I'm familiar with have settled on the number of dead as 177
million, although I've seen estimates of up to 200 million.
-Bob Wallace
I did find the simulator, on the Franz Examples and Utilities page, at:
http://examples.franz.com/index.html
I haven't looked at the CMU repository in a long time though.
Thanks,
--Bruce
Fred Gilham wrote:
> Sashank Varma wrote:
>
>>The code was available at the CMU repository last I checked.
>
>
> If you're talking about the *Lisp simulator, J. P. Massar has
> maintained that code, and a year or so ago released an updated
> version --- f20.
>
On 31 Aug 2004 08:16:01 -0700, Fred Gilham
<······@snapdragon.csl.sri.com> wrote:
>
>Sashank Varma wrote:
>> The code was available at the CMU repository last I checked.
>
>If you're talking about the *Lisp simulator, J. P. Massar has
>maintained that code, and a year or so ago released an updated
>version --- f20.
I don't know where one could find the source code for the actual
CM 2 software (which would include the low level code that did the
talking between the host and the CM, the implementation of PARIS
(the CM2 assembly language), the source for *Lisp, for C* and other
stuff.)
I would bet, though, that someone, somewhere, has backup tapes
in their attic or basement or drawer...
I'd be happy if I cound find some distribution of the CM software for
the Symbolics machines. I don't really need the source itself.
Thanks,
--Bruce
JP Massar wrote:
> On 31 Aug 2004 08:16:01 -0700, Fred Gilham
> <······@snapdragon.csl.sri.com> wrote:
>
>
>>Sashank Varma wrote:
>>
>>>The code was available at the CMU repository last I checked.
>>
>>If you're talking about the *Lisp simulator, J. P. Massar has
>>maintained that code, and a year or so ago released an updated
>>version --- f20.
>
>
>
> I don't know where one could find the source code for the actual
> CM 2 software (which would include the low level code that did the
> talking between the host and the CM, the implementation of PARIS
> (the CM2 assembly language), the source for *Lisp, for C* and other
> stuff.)
>
> I would bet, though, that someone, somewhere, has backup tapes
> in their attic or basement or drawer...
>
>
On Tue, 31 Aug 2004 21:24:56 -1000, Robert Bruce Carleton
<···@hakuhale.net> wrote:
>I'd be happy if I cound find some distribution of the CM software for
>the Symbolics machines. I don't really need the source itself.
>
CM installations using Symbolics machines as front ends always came
with source, IIRC.
So, if you could locate a distribution tape, it would have the source
on it. The trick would be finding a distribution tape / cartridge or
whatever we used back then.
ok, I get it. Thanks for clearing that up.
--Bruc
JP Massar wrote:
> On Tue, 31 Aug 2004 21:24:56 -1000, Robert Bruce Carleton
> <···@hakuhale.net> wrote:
>
>
>>I'd be happy if I cound find some distribution of the CM software for
>>the Symbolics machines. I don't really need the source itself.
>>
>
>
> CM installations using Symbolics machines as front ends always came
> with source, IIRC.
>
> So, if you could locate a distribution tape, it would have the source
> on it. The trick would be finding a distribution tape / cartridge or
> whatever we used back then.
>
>