Author: Uri Blass
Date: 13:40:16 11/20/03
Go up one level in this thread
On November 20, 2003 at 16:21:56, Russell Reagan wrote: >On November 20, 2003 at 15:27:33, Uri Blass wrote: > >>Longer time control help the stronger programs. > >That is not true. A longer time control helps the program with more weaknesses, >because it has more time to search deeper and cover up its weaknesses. This is >why we often hear people say that "engine X is very strong at long time >controls", but at shorter time controls they don't say that, because its >weaknesses get exposed and it isn't as strong. I have played lots of test games >at different time controls, and looked at the results of other people's >tournaments where different time controls were used, and this is the trend I >have seen in almost all cases. I am not using this as an excuse to losing on equal hardware but talk about games with unequal hardware. Try Crafty against weaker winboard engine like gnuchess when you give the weaker engine big hardware advantage. Crafty is going to lose at short time control and win at long time control. This experiment was already done by people. The same was done with Crafty against Noonian chess. I also did the same experiment with old movei against tscp when I told movei to play 5 times faster and movei lost at fast time control and won at longer time control. Uri
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.