Author: Uri Blass
Date: 08:23:41 12/31/04
Go up one level in this thread
On December 31, 2004 at 11:03:07, Torstein Hall wrote: >On December 31, 2004 at 09:58:04, Uri Blass wrote: > >>On December 31, 2004 at 09:27:00, Ziad Haddad wrote: >> >>>According to my own tests, Shredder 8 is slightly than Shredder 7.04. The mai >>>difference according to me is due to the fact that Shredder8 ponders 15% fastly >>>than his predecessor. But i've got a great assumption, always with my own tests >>>i find that Shredder7.04 with my Haddad1.2DP2 setting is stronger than >>>Shredder8. When facing one each other Shredder 8 beats my setting but the >>>results of matches vs the other top engines are slightly in favor of >>>Shredder7.04. That's why i'm actually trying to find a better setting for >>>Shredder8. >>>I wonder if SSDF will test my setting as they done with CM9000 SKR then Shredder >>>7.04 will be on top again. >>>The latest SSDF rating prooves at least something: Stefab Meyer Kahlen is the >>>beggest computer chess programmer of the last century. Shredder will certainly >>>remain on top for a few years again. >> >>We cannot know the future. >>I wonder how do you know what is going to happen in the next few years. >> >>When I consider the fact that movei is only 200-300 elo weaker than shredder >>inspite of all the missing knowledge that it does not have(both in evaluation >>and search) > >I think the improvments get more and more difficult as you move up in strength >elovise. I do not totally agree here. I believe that improvement is easier to get after you write the relavant tools to check if you have bugs and I still did not write the relevant tools. Improvement is also easier to get when you have more free source code to learn from it. More parameters to tune, more parameters kicking in at wrong time etc. >etc. I think that the current chessprograms are strong, but still with a lot of >potential for improvment. But have we really seen any real improvment in 2004? > >Torstein Of Course we saw an improvement. For example Fruit became significantly better. The improvement was not of the top programs but my opinion is that the programmers of the top programs are not the best and their main advantage is simply that they worked for more time about their program. 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.