Author: Robert Hyatt
Date: 16:25:17 12/20/99
Go up one level in this thread
On December 20, 1999 at 16:11:10, Greg Lindahl wrote:
>On December 17, 1999 at 22:58:36, Dann Corbit wrote:
>
>>There is a recent discussion on the computer chess club internet forum about the
>>feasibility of creating a chess chip using FPGA. Now, a chess chip *must* have
>>very fast access to large amounts of shared memory, and also be able to execute
>>an instruction set (much like a general purpose CPU -- but the instruction set
>>is {of course} specialized for chess).
>
>I'm not so sure that people would consider Deep Blue's chips to have either a
>"large amount of shared memory" or an instruction set much like a general
>purpose CPU. Changing your assumptions changes your answer, of course.
He is talking about the hash table. If you punt on that, you take a huge hit
in performance. The only way to make it up is to do dozens of chess processors
like DB used. This is assuming that the chess processor is really doing a true
alpha/beta search, and not just pieces of this (like evaluation).
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.