Search found 19035 matches

by hgm
Wed Mar 05, 2008 6:00 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Extreme late move reductions?
Replies: 8
Views: 3836

Re: Extreme late move reductions?

What I would like to try is to start with LMR of 2 ply on every move, but with the window opened to about 1 pawn below the current evaluation. Moves that fail low w.r.t. this threshold then stay at this depth, moves that get above it are re-searched with an LMR of one against the original alpha.
by hgm
Sun Mar 02, 2008 7:17 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Repetition detection structure.
Replies: 194
Views: 206702

Re: Repetition detection structure.

An honest question for hgm. I've seen a lot of posts from you written in this adversarial style, and I'm just wondering, why do you write like this? If the goal is to discuss things productively and share ideas and learn from each other, what is gained by the adversarial tone? It seems like most of...
by hgm
Sun Mar 02, 2008 10:20 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Repetition detection structure.
Replies: 194
Views: 206702

Re: Repetition detection structure.

For instance you said: "As the table is so small that it always resides in the L1 cache, this must be faster than the linear-array method." which presumes that the linear array doesn't fit in the cache or that in the process of searching, one would have to hit L2 or main memory. The evidence presen...
by hgm
Sat Mar 01, 2008 11:06 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Repetition detection structure.
Replies: 194
Views: 206702

Re: Repetition detection structure.

I decided now to look at Joker's games and here is one game that Joker lost Thanks for looking at the games, Uri. I am currently so busy I haven't had time to do that yet. The Cheese game seems no 50-move case to me. It is Joker there that finally decides to make the capture, and Joker already had ...
by hgm
Sat Mar 01, 2008 10:42 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Repetition detection structure.
Replies: 194
Views: 206702

Re: Repetition detection structure.

Let me rephrase my comment. "What exactly are you talking about?" Well, read back and you will know... In short: you are claiming that 1/x is always smaller than 10/y because 1 < 10, no matter what x and y are ("regardless of investment"). That says much for your abilities in arithmetic. As for the...
by hgm
Sat Mar 01, 2008 10:15 am
Forum: Computer Chess Club: Tournaments and Matches
Topic: Sloppy experiment, results after 1 cycle
Replies: 8
Views: 1922

Re: Sloppy experiment, results after 1 cycle

I probably missed that question, and I must admit I am not sure I understand what exactly you are asking now. Do you mean what the minimum number of games is that an engine should have before you can calculate its rating? I think this is 1 game. If the calculated rating means anything is another mat...
by hgm
Sat Mar 01, 2008 8:26 am
Forum: Computer Chess Club: Tournaments and Matches
Topic: Sloppy experiment, results after 1 cycle
Replies: 8
Views: 1922

Re: Sloppy experiment, results after 1 cycle

Yes, you see how dangerous it is to do only one-sided testing. It is really important ot have both weaker and stronger opponents. Some engines have a much wider distribution of performance than the standard Elo curve assumes. (Joker also seems to suffer from this.) Such engines look poor if you test...
by hgm
Sat Mar 01, 2008 8:16 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Chess Tournaments
Replies: 4
Views: 1936

Re: Chess Tournaments

Yes, and there are actually two Dutch tournaments: the Open Dutch Computer-Chess Championship (ONK), usually early November, and a tournament where no title can be earned (ICT), usually in May.

I also heard of the Polish Open.
by hgm
Sat Mar 01, 2008 7:53 am
Forum: Computer Chess Club: General Topics
Topic: Time-Odds tournament
Replies: 43
Views: 7422

Re: Time-Odds tournament

Perhaps some explanation: The basic tournament with 23 engines was finished, but failed to accomplish one of its goals, namely to do more symmetric testing on ArcBishop80, BigLion80 and Chancellor (i.e. not only against stronger opponents). So I decided to add a few more strongly handicapped version...
by hgm
Sat Mar 01, 2008 7:32 am
Forum: Computer Chess Club: Tournaments and Matches
Topic: Joker in my Division Junior now on 3rd
Replies: 13
Views: 2626

Re: Joker in my Division Junior now on 3rd

And I haven't really worked on it for half a year now! :shock: (Since Open War 3 started, and that version was a bust, which I now retracted.) Is this blitz? I noticed that Joker is much stronger (compared to other engines) at blitz than at long time controls. The most likely explanation is that it ...