Author: Dann Corbit
Date: 13:22:03 12/20/05
Go up one level in this thread
On December 20, 2005 at 14:39:49, Ernest Bonnem wrote: >Using an automatic testing of test-suites with Rybka will only give you an >approximation for the results. >The only way to obtain reproducible results (until Vas corrects that small bug) >is to exit the engine and launch it again after each position test! >You can check that easily for yourself... > >So Mike may try as hard as he can, he will never get your results. So there is some element of randomness due to a hash bug. Even at that, Rybka solves more of these positions than other engines do. >By the way, saying that your A64 2.2GHz is only 67 % as fast as Mike's machine >(1.8GHz Opteron) seems a bit harsh! That is the ratio when he as a 64 bit OS and I don't.
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.