Author: Ed Schröder
Date: 23:54:24 09/26/99
Go up one level in this thread
>Posted by Fernando Villegas on September 26, 1999 at 17:44:39: > >RC has been improved in his speed in some susbtantial amount in middle and >ending parts of the game, as the Rebel site advertise. Question is, Ed: how >much this means in terms of playing strenght? My fastest equipment is 266 Mhz. >What Can I wait from RC with the last minute change? >Fernando The last-minute speed improvement came from a discussion here in CCC. The discussion started with the subject "mate-values in hash tables" but then the discussion changed to the subject about updating the best-move in the hash table mostly by Bob and me. As a result I was able to make Rebel 8% faster by just adding one instruction. Wish I had such days (say) 10 times a year :) From one thing came another, I had a critical look at move ordering again, made some changes, did intensive testing and there it was: 10-15% faster in the end-game and 25-30% faster in the middle-game. After 18 years wrestling, fiddling and twiddling with move-ordering this is remarkable as I expected move-ordering was close to perfect in Rebel. I wonder how much improvement there is still left in move-ordering. The nice thing working on move-ordering improvements is that there is no risk of making mistakes. You don't change the chess knowledge of the program, you only change the order of the moves being searched. As a result you can make the program slower or faster. If you manage to make your program faster it's a free non-risk improvement valid for all possible PC configurations including the PC's that will come in the future. I don't know what a 25% speed improvement means in terms of playing strength. Opinions are divided. I would say it gives 5-10 elo points. Others will say 15-20 and they can be perfectly right too. Ed
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.