Author: Uri Blass
Date: 08:05:38 12/24/02
Go up one level in this thread
On December 24, 2002 at 06:06:41, Rafael Andrist wrote: >On December 23, 2002 at 05:46:26, Uri Blass wrote: > >>I did not learn most of the source code in crafty and I am afraid that if I try >>to change something that I do not understand I may have bugs(I have bugs even >>when I try to change something that I understand). > >I also did not learn the code of crafty, I only looked at the EGTB and the >Nullmove implementation and nothing more. But I had no problems to understand >these things and even without knowing the complete context. I guess your ideas >are mainly search-related and therefore they should not be too difficult to >implement - you don't need to know how Crafty's evaluation and move generation >works in detail if you do that. I have ideas that are search related. My main problem is not to think about ideas but to learn things that other did. I think that it may be productive to have examples of changes that are not productive to the source code so people with productive ideas can learn how to change the code. I also need to know about how to compile it correctly because I understood that something else except normal optimize for speed was done that I did not care to learn or to buy. > >>I guess that I may get more positive responses also if I improve movei and make >>it better in tactics than crafty(It is already better than crafty in many >>positions inspite of not using hash tables in an efficient way). > >It is always encouraging when the program is better in some positions, but this >doesn't mean anything about the real playing strength. I know and people like vincent may claim that I use wrong positional pruning. I already got 2 responses so I will see(today I still prefer to try to improve movei and not to learn about changing crafty) 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.