Author: Uri Blass
Date: 11:08:02 12/17/04
Go up one level in this thread
On December 17, 2004 at 13:56:30, Dann Corbit wrote: >On December 17, 2004 at 06:07:53, Uri Blass wrote: > >>On December 17, 2004 at 05:55:47, Joachim Rang wrote: >> >>>On December 17, 2004 at 05:48:56, Uri Blass wrote: >>> >>>>Looking at the first game I also find depth 2 moves. >>>> >>>>Uri >>> >>>As a betatester I can confirm, that Fruit is still no match for Shredder. It is >>>probably at least 150 elopoints weaker. the upcoming Fruit 2.0 (around >>>christmas) will be stronger than 1.5 but still far behind the commercial ones. >>> >>>regards Joachim >> >>It is clear that it is too early for it to be better than shredder7.04 >>I expected it to be better than shredder7.04 and even better than shredder8 but >>my prediction was for 2007 and not for today. > >I watched the first three games. Most of the time, Shredder thought for quite a >while, just like Fruit did. In most of the moves it could search more than 2 plies but some consecutive moves at depth 2 are enough to decide the game. > >Perhaps there is some Shredder/Arena related bug. One thing is sure, that Fruit >2 will be strong. My other programs do not perform so well against Shredder >under Arena. I do not know If you play with ponder off then the bug may be also a bug of fruit that does not let shredder to use processor time. Another possibility is that shredder is slowed down because of using too much hash[I think that it may be a good idea to use only 2/8 of the memory for hash (1/8 for every program) to avoid problems and it means that if you have 256 mbyte Shredder should use 32 mbytes for hash like fruit and if you have 512 mbytes use 64 mbytes for shredder and 64 mbytes for fruit]. 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.