Author: Andrew Slough
Date: 14:56:48 11/08/99
Go up one level in this thread
On November 08, 1999 at 17:24:27, Bruce Moreland wrote: >On November 08, 1999 at 15:26:47, Andrew Slough wrote: >>Stephen Streater had a program a while back that ran on a 200Mhz StrongARM >>(which is single issue, 32 bit), running on a 16Mhz memory bus that did over >>2000kNPS. The ARM architecture is _really_ nice for chess. >> >>Andy > >Did it play or was it a test bed? It's easy to get a man to the moon if you are >allowed to distribute him over an area several miles square. > >bruce It actually played, but it wasn't very good. It was a bit like TECH in its positional evaluation :-) The posts from a few years ago are on dejanews. It was fast because the whole board was stored in registers and one instruction on the ARM could extract the type of piece + the colour and set condition flags based on that. It was full alpha-beta with move ordering and hashing etc, but he doesn't believe in null-move. Andy
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.