Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Tiger against Deep Blue Junior: what really happened.

Author: Robert Hyatt

Date: 19:16:28 07/26/00

Go up one level in this thread


On July 26, 2000 at 12:01:40, Chris Carson wrote:

>On July 26, 2000 at 11:48:25, Albert Silver wrote:
>
>>On July 26, 2000 at 01:35:40, Ed Schröder wrote:
>>
>>>On July 25, 2000 at 19:45:11, Dave Gomboc wrote:
>>>
>>>>On July 25, 2000 at 17:29:34, Ed Schröder wrote:
>>>>
>>>>>On July 25, 2000 at 16:54:39, Alvaro Polo wrote:
>>>>
>>>>>>I worked for IBM as a scientist at the IBM Scientific Center in Madrid. I would
>>>>>>very much more trust Hsu's number than "official IBM" numbers. PR's and
>>>>>>marketers at IBM are not stupid people (my father was a country general manager
>>>>>>there), they are on the contrary very intelligent, but they don't care that much
>>>>>>about scientific exactness in documents directed to the general public. They
>>>>>>probably wouldn't understand very well, for example, why the difference between
>>>>>>256 and 480 processors is significant.
>>>>>>
>>>>>>Alvaro
>>>>>
>>>>>With all respect to your opinion I believe that P/R people very well
>>>>>understand the value of numbers. If they don't they would do a very
>>>>>poor job which I find hard to believe.
>>>>>
>>>>>Ed
>>>>
>>>>That was quite a statement from Alvaro. :)
>>>>
>>>>In any case, DB2 had 480 chess processors, not 256.
>>>>
>>>>Dave
>>>
>>>Sigh. From the IBM pages again:
>>>
>>>    "The latest iteration of the Deep Blue computer is a 32-node
>>>     IBM RS/6000 SP high-performance computer, which
>>>     utilizes the new Power Two Super Chip processors
>>>     (P2SC). Each node of the SP employs a single
>>>     microchannel card containing 8 dedicated VLSI chess
>>>     processors, for a total of 256 processors working in
>>>     tandem. The net result is a scalable, highly parallel system
>>>     capable of calculating 60 billion moves within three minutes,
>>>     which is the time allotted to each player's move in classical
>>>     chess."
>>>
>>>It says 256 processors. The URL:
>>>
>>>http://www.research.ibm.com/deepblue/meet/html/d.3.2.html
>>>
>>>Then look at the logo, it says the re-match. So 256 processors.
>>>
>>>Ed
>>
>>Well, according to the horse's mouth, these are the very first lines (including
>>the title) of his article published in IEEE/1999:
>>
>>"IBM?S DEEP BLUE CHESS
>>GRANDMASTER CHIPS
>>
>>THE IBM DEEP BLUE SUPERCOMPUTER THAT DEFEATED WORLD CHESS
>>CHAMPION GARRY KASPAROV IN 1997 EMPLOYED 480 CUSTOM CHESS CHIPS.
>>THIS ARTICLE DESCRIBES THE DESIGN PHILOSOPHY, GENERAL ARCHITECTURE,
>>AND PERFORMANCE OF THE CHESS CHIPS, WHICH PROVIDED MOST OF DEEP
>>BLUE?S COMPUTATIONAL POWER."
>>
>>                                Albert Silver
>
>Albert,
>
>Did the article discuss NPS?  I do not think it matters, but I am
>interested in what it says (NPS vs NPS means nothing).  Also, which
>IEEE journal is this, I assume the publication that goes to all members,
>however, there are several IEEE societies and each has a different
>tech journal.  Do you have the month?  Thanks in advance if you can
>supply this information.  I will be a the library later this week doing
>part of a literature search and I may try to find this article (during
>a break).
>
>Best Regards,
>Chris Carson


I _believe_ it was IEEE Micro, March 1999.  But I am not at the office and am
not sure.  Don't go ballistic if that is wrong...



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.