Search found 211 matches

by asanjuan
Thu Jul 19, 2012 9:30 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Aspiration Windows: Rubbish!
Replies: 56
Views: 13833

Re: Aspiration Windows: Rubbish!

1: first case is exactly what you describe 2: second case, if this node is started with aspiration bounds and there might be no "score > alpha", or no other condition that leads to a real value. (so the complete move loop may be pruned ??!) and additionally no bestValue update in a failsoft framewo...
by asanjuan
Wed Jul 18, 2012 1:21 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Aspiration Windows: Rubbish!
Replies: 56
Views: 13833

Re: Aspiration Windows: Rubbish!

You can futility-prune in a PV node after alpha is raised, right? The window is only {-INF, INF} on the search of the first move. After that alpha will be raised to some finite value, which very well might be above the current eval. E.g. suppose the first move captured a hanging Queen. After that, ...
by asanjuan
Wed Jul 18, 2012 10:50 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Aspiration Windows: Rubbish!
Replies: 56
Views: 13833

Re: Aspiration Windows: Rubbish!

I have tested and tested, and I have one conclusion. They simply don't work for me. I've watched many a test game where it searches normally, and is doing quite well, then suddenly gets a 'persistent phantom mate score' which pushes the window size to the point where it takes 5 seconds to perform a...
by asanjuan
Fri Apr 27, 2012 8:27 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Error bar calculations - help!
Replies: 11
Views: 2931

Re: Error bar calculations - help!

I alway thought that the relevant point is that you choose the poorest individual to die, and among the bests for reproduce. This way we improve the average strength of the population. well, there are a tons of different G.A. It depends on how you want your population to be, but the idea of create ...
by asanjuan
Fri Apr 27, 2012 8:01 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Error bar calculations - help!
Replies: 11
Views: 2931

Re: Error bar calculations - help!

The sources to bayeselo are available if you are willing to dive into that. I wasn't looking for sources, but for a simple explanation of the ecuations. So the links provided by Fermín and Jesus were very helpful. But on the other hand... Dan, i'm very impressed with your response. The algorithm I ...
by asanjuan
Wed Apr 25, 2012 7:58 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Error bar calculations - help!
Replies: 11
Views: 2931

Error bar calculations - help!

Does anybody know where can i find a paper that describes how to calculate the error bars for elo in a tournament?
I'm trying to improve my G.A with tournament play, but taking into acount the error margin to build a more precise fitness function.

Maybe Remi or Jesus muñoz can give me more info.
by asanjuan
Wed Apr 04, 2012 10:06 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: New engine: Rhetoric
Replies: 13
Views: 3550

Re: New engine: Rhetoric

It still loses against Eveann, but as we say: "ya tenía ganas de publicarlo".
by asanjuan
Wed Apr 04, 2012 8:07 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: New engine: Rhetoric
Replies: 13
Views: 3550

Re: New engine: Rhetoric. GA papers.

Sounds very interesting - I was wondering how GAs work in a discreet problem space. Is it public how it internally works - how the genotype looks like, mutations, crossovers, etc? Rhetoric is based on this article. You can download it from my site: http://www.chessrhetoric.com/index.php/downloads/v...
by asanjuan
Tue Apr 03, 2012 2:18 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: New engine: Rhetoric
Replies: 13
Views: 3550

New engine: Rhetoric

Hello, I would like to present the first release of my own engine: Rhetoric. Not a very strong engine, because is in the 2200 - 2300 elo range, but is surprisingly strong playing against humans. But what makes Rhetoric special is the fact that it has been trained with Genetic Algorithms trying to fi...
by asanjuan
Mon Feb 06, 2012 12:12 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Moving to Magic Bitboards... any advice?
Replies: 14
Views: 3424

Re: Moving to Magic Bitboards... any advice?

Ok, i have used a profiler. I didn't know its existance because my VS2005 is in spanish, and here it seems to be known as "instrumentación". After analyze my code, i realized that my engine spends 36% of the time in eval. This is what i knew by intuition... the only difference is that i have real nu...