Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: HW based Crafty (Boule's thesis)

Author: Robert Hyatt

Date: 10:25:59 04/04/02

Go up one level in this thread


On April 04, 2002 at 04:07:42, Tom Kerrigan wrote:

>On April 03, 2002 at 17:49:58, Robert Hyatt wrote:
>
>>Yes I am aware of what an FPGA is.  I believe that a SOTA (state of the art)
>>fpga can run just as fast as the 3 micron (three micron, not .3) micron process
>>ASICS used in the original DB hardware...  3 micron ASICS in DB only ran at
>>24mhz.  An FPGA can certainly be run at that clock speed...
>
>The information I found on the web indicates that Chiptest, not DB, was done
>with 3 um features. And that the DB or DB2 chips had 0.6 um features. (There is
>some ambiguitiy.) It makes sense that the DB and DB2 chips have the same feature
>size, though. They ran at pretty much the same clock speed, and I don't remember
>reading about FHH doing a shrink.
>

I didn't write very clearly.  Chiptest and deep thought (the original deep
blue hardware) was, I believe, 3 micron.  DB1 and DB2 were complete re-designs
using smaller die sizes.  This is mentioned in his book, when/if it comes
out...




>I know that simple, pipelined, carefully floorplanned, general-purpose CPU
>designs run at ~100MHz on modern FPGAs. From what I know of the
>Belle/Chiptest/DT/DB move generator, I'd expect its critical paths to be at
>least twice as long, and probably longer. So I'd put an optimistic upper bound
>of 50MHz on it. And that's just the move generator; I'd expect the eval logic to
>be slower. But if you take the optimistic 50MHz number and the very overly
>optimistic 2-cycle-per-node number, you get 25M NPS, not "30M NPS easily."

All I really know about the DT chip was that it took 10 cycles per node...
although some cycles were skipped (fast vs slow eval for example)...  I based
the 30M nps value on statements made by Hsu...




>
>>I'm only quoting from the title of a paper Ken wrote, which was "An FPGA chess
>>move generator" or something to that effect (I quoted the exact title elsewhere
>>in this thread...)  I think you can locate a reference to it by searching for
>>"Belle chess machine" on Google...
>
>The only paper I can find that's remotely similar to what you're talking about
>is Marc Boule's paper, from June, 2001. Even if Ken wrote a paper with that
>title, it doesn't mean the paper was written about something that was done in
>Belle.
>
>-Tom


Here is the reference from Google:

[DOC] An FPGA Based Move Generator for the Game of Chess
... Condon and K. Thompson, ?BELLE Chess Hardware?, 1979


It was published in 1979.  Implies to me that "FPGA" means just what it
should mean, and this was the circuit used in the 1977 Belle machine (this
was searching at 5K nodes per second when chess 4.7 was searching at 2.3K
nodes per second on the fastest CDC machine available at the time, the Cyber
176).

It seems to be a description of his hardware move generator stuff that he
used in that version of Belle.  1980 Belle was a full-blown engine-in-hardware
design and searched around 160K nodes per second...



This page took 0.02 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.