TECHNET Archives

October 2006

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Steve Gregory <[log in to unmask]>
Reply To:
TechNet E-Mail Forum <[log in to unmask]>, Steve Gregory <[log in to unmask]>
Date:
Mon, 9 Oct 2006 15:27:42 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (230 lines)
WOW! Simply AMAZING! Inge has sent me pictures of the wafers:

http://stevezeva.homestead.com/files/PICT0018.jpg

There's only been once in my life that I've been involved in something you might call "supercomputers". It was back in San Jose, California around 1989 or thereabouts.

I was working for Comptronix (no longer around) and we built two of the boards that were a part of a MasPar Computer Corporation (also no longer around) MP-1 mini-supercomputer. The two boards that we built was the Processor Element (PE) assembly, and the Array Control Unit (ACU) assembly. Both boards were 24" X 24". The PE had 64 processors (custom 180-pin QFP's) on it which wasn't too bad because they were all the same part number, but the ACU had 115 different programmed devices on it, most of them were PLCC-20's and 28's. That was in addition to about 45 other part number IC's, and about 30 passive component part numbers.

We only had a Fuji CP-4 and FIP-2 at that time...we didn't have enough room for feeders to load everything on the ACU in one pass, so we first ran the boards in two passes which was pretty cumbersome. Then I came-up with the idea to sequence all the PAL's in "PAL Sets" for each board and run them in one feeder on the FIP-2. We had to run the pick and place machine in error stop in case there was a vision or pick-up error, we could recover right back at the same step, otherwise you could load the rest of the board wrong...

I was pretty impressed with what I learned about MasPar's computers, but they don't hold a candle to what you were showing us! I hope Dr. Carlstedt starts his project back up!

I think we should do this again...that is having this contest. It was pretty entertaining trying to figure out what we were seeing!

-Steve-   

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Hfjord
Sent: Monday, October 09, 2006 1:51 PM
To: [log in to unmask]
Subject: Re: [TN] NTC: Friday question 2

Dear Yoda,

I was a little involved in a secret computer project 1992. The inventor, doctor Carlstedt of Chalmers University, in cooperation with MIT, Stanford, Plessey and lots of others, had invented a quite new programming language (still not available to publicum) which simplified the way computers work.
He was decades in advance. By means of hired software guys from India and the US, he created hardware for this purpose. The computer is not a common paralell working type, but uses something named 'closures', which means that instead of storing ordinary 1,0 bits, it stored whole words. These words are floating, and other processors on the wafer use these words by means of association. Each wafer had 500 powerful processors, each one with higher capacity than the fastest PC. The prototype consisted of 10 wafers stacked upon each other, with vertical connections. The wafers were put into something like a CD drive (not rotating of course), so you could easily take out a wafer and put in another one. The dissipation heat was incredible when the computer worked at most, something like 12 kW! And the computer was not bigger  than a thick book! Watercooled. Several such modules could be paralelled without limitation. A few blocks had bigger capacity than a Cray.
Unfortunately, the investor got more interested in other areas, which gave more money on short time, and they did not believe that dr Carlstedt's computer could compete with Intel, Cray, IBM and others (lack of competence, said dr Carlstedt) so the whole project was stopped. He got just a couple of years from start to finished product, then they wanted money back.  But the story is not ended, dr Carlstedt just refines his project and is awaiting contract time to run out, then he can start it all over again.

I'm not a computer guy, so I can't judge anything, but just recapitulate what I'm told, but I must say it was a very, very interesting project. Dr Carlstedt has withdrawn, he has bought a farm and lives a now quiet life together with his wife, far from the heat.  Who knows, one day he may come out of his igloo and make a try again. Besides computer interests, he works hundreds of hours a year with archeology and a project to protect eagles.
He has lots of patents too, over 100 I think he said.

The wafers aren't like anything else in this world, they are extremly complicated and I was allowed to present pictures, because Dr Carlstedt said that noone will understand nothing at all. However, I was not allowed to show high magnification images.

That's it.

Inge




----- Original Message -----
From: "Steve Gregory" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Monday, October 09, 2006 4:33 PM
Subject: Re: [TN] NTC: Friday question 2


Definitely not a Cray, here's a picture of a Cray Processor board:

 http://stevezeva.homestead.com/files/Processor_board_cray-1_hg.jpg

Has to be something much faster. Could it be the processors for AaAP (Asynchronous Array of simple Processors)?

>From the following link:
>http://en.wikipedia.org/wiki/Asynchronous_Array_of_Simple_Processors#As
>AP_1.0_Chip

"The Asynchronous Array of simple Processors (AsAP) architecture comprises a 2-D array of reduced complexity programmable processors with small memories interconnected by a reconfigurable mesh network. AsAP was developed by the VLSI Computation Laboratory (VCL) at the University of California, Davis and achieves high performance and high energy-efficiency, while requiring a relatively small circuit area.

AsAP processors are well suited for implementation in future fabrication technologies, and are clocked in a Globally Asynchronous Locally Synchronous
(GALS) fashion. Individual oscillators fully halt (leakage only) in 9 cycles when there is no work to do, and restart at full speed in less than one cycle after work is available. The chip requires no crystal oscillators, PLLs, DLLs, or any global frequency or phase-related signals whatsoever.

The multi-processor architecture efficiently makes use of task-level parallelism in many complex DSP applications, and also efficiently computes many large tasks utilizing fine-grain parallelism."


-Steve-

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Ingemar Hernefjord
(KC/EMW)
Sent: Monday, October 09, 2006 8:51 AM
To: [log in to unmask]
Subject: Re: [TN] NTC: Friday question 2

Not a Cray, but processing and memory is OK. /Inge

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of David D. Hillman
Sent: den 9 oktober 2006 14:47
To: [log in to unmask]
Subject: Re: [TN] NTC: Friday question 2

Hi folks!  Charlie was actually pretty close in guessing the identity of the "mystery creature" pix Steve posted from me on Friday. You are looking at calcium, sodium, and various other crystal structures that formed as a droplet of water based cleaning solution evaporated!

Inge - is your pix a portion of a Cray computer processing and/or memory die?

Dave



             Charlie Barker
             <charlie.barker@S
             UNTRONICINC.COM>                                           To
             Sent by: TechNet          [log in to unmask]
             <[log in to unmask]>                                          cc

                                                                   Subject
             10/06/2006 09:24          Re: [TN] NTC: Friday question 2
             AM


             Please respond to
              TechNet E-Mail
                   Forum
             <[log in to unmask]>
             ; Please respond
                    to
              Charlie Barker
             <charlie.barker@S
             UNTRONICINC.COM>






I don't know about Inge's pix, but Dave has somehow captured a frost upheaval from my freezer!

Charlie B.

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Steve Gregory
Sent: Friday, October 06, 2006 8:30 AM
To: [log in to unmask]
Subject: Re: [TN] NTC: Friday question 2

Okay Everybody, here's Inge's mystery photo:

http://stevezeva.homestead.com/files/PICT0013d.jpg

I don't have a clue what it is yet...

Also Dave Hillman sent me a picture to post for our Friday guessing
contest:

http://stevezeva.homestead.com/files/Unknown_Creature.jpg

I don't have a clue about that one either...

-Steve Gregory-

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Ingemar Hernefjord
(KC/EMW)
Sent: Friday, October 06, 2006 3:26 AM
To: [log in to unmask]
Subject: Re: [TN] NTC: Friday question 2

Sending Friday picture to Steve.
This is probably a very hard to guess one, so, I will send another photo, and still another, until the coin falls for someone.
Inge


-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Hfjord
Sent: den 5 oktober 2006 22:15
To: [log in to unmask]
Subject: [TN] NTC: Friday question 2

Hi,
seems as a little contest was appreciated. Unfortunately, I don't have time to send picture in time to Steve's wonderland screen, will be posted later.
However, I'll give you a thread. The real genious will find out what I'm going to show. The thread is this:  H

Good Luck¨
Inge

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or
(re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or
(re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or
(re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16
for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or
(re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16
for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or
(re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or
(re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16
for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------

ATOM RSS1 RSS2