Author: Robert Hyatt
Date: 14:50:00 12/06/01
Go up one level in this thread
On December 06, 2001 at 11:20:32, Severi Salminen wrote: >>I ran the test and you are absolutely correct. 32 bit pawn hash keys >>are not usable. I am not sure when I went to this, but I did look back >>thru my test results in my file at the office, and the last time I did any >>hash sanity testing was when I did use 64 bits. Since I made the change to >>32 I apparently didn't test well enough. >> >>It is a trivial change to fix, and Crafty is now back to 64 bit signatures. >> >>Glad you took the time to test and kept the idea "alive". You were >>certainly correct about the number of collisions... and it has to be >>unacceptable. > >So how many errors are you getting (in form of errors/searched nodes) with >32-bit keys? And no errors with 64-bit keys? > >Severi No errors with 64 bit keys. What I did was have crafty "annotate" several games so that it would pass over opening, middlegame and endgame positions. With 64 bits the "debug test" never complained. With 32 bits, it was complaining every 2-4 seconds, this on a single-cpu PIII/750 laptop. This machine searches around 300K normally, so that says that I was seeing a collision every 500K-1M nodes roughly. It wasn't always that frequent, and it was sometimes much more frequent. Very "bursty" at times, but overall the average was above what I consider to be reasonable and I am back to 64 bit signatures (2 line change in my code to fix this).
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.