Search found 3101 matches

by Guenther
Mon Jan 16, 2012 9:28 am
Forum: Computer Chess Club: General Topics
Topic: Another strong engine
Replies: 65
Views: 16378

Re: Another strong engine

The beta copy of Matrix 3000 that I received is a clone of Houdini 1.5a x64. I did a binary compare and I only see a few changes like program name and some UCI options (Hash , Probe Depth). It still lists Robert Houdart as the author :evil: :evil: :evil: True. A hex edited Houdini 1.5a So you told ...
by Guenther
Wed Jan 11, 2012 8:28 pm
Forum: Computer Chess Club: General Topics
Topic: Rybka Investigation : Survey Of Evidence
Replies: 77
Views: 13239

Re: Rybka Investigation : Survey Of Evidence

there are still millions who preach fanatically that the world and species were created in 7 days , despite overwhelming lack of evidence and common sence Are you familiar with programming, especially with chess programming? And just in case you aren't, do you think that it is up to non-programmers...
by Guenther
Sat Jan 07, 2012 9:41 pm
Forum: Computer Chess Club: General Topics
Topic: Rick48 going commercial
Replies: 24
Views: 7072

Re: Any surprise !

Surprise :o http://rick48.com/ Jouni Hi ! I heard about this engine from many years. Maybe I have also an old copy ! A Polish friend connexion ! Take a look here -please- (the number four): http://wbec-ridderkerk.nl/html/BayesianElo_ed18.htm http://chessprogramming.wikispaces.com/Stanislaw+W.+Rostk...
by Guenther
Sun Dec 04, 2011 10:44 am
Forum: Computer Chess Club: General Topics
Topic: methods of testing engines
Replies: 25
Views: 2707

Re: methods of testing engines

Using these 2-ply positions is somewhat wasteful, as some of the positions probably give about 85% White expectancy, but as you say with reversal of colors the test is still valid. Elo difference will be distorted towards zero, but again it will not cause the weaker engine to come out on top. The b...
by Guenther
Sat Dec 03, 2011 9:41 pm
Forum: Computer Chess Club: General Topics
Topic: methods of testing engines
Replies: 25
Views: 2707

Re: methods of testing engines

Anyway, your fuller description of your starting positions makes my question unnecessary. I see nothing wrong with your method. Adam Actually there is a lot wrong with his method if he wants to measure strength or something similar between programs. I also used a full2ply set once, but only for exp...
by Guenther
Sat Dec 03, 2011 6:15 pm
Forum: Computer Chess Club: General Topics
Topic: methods of testing engines
Replies: 25
Views: 2707

Re: methods of testing engines

Anyway, your fuller description of your starting positions makes my question unnecessary. I see nothing wrong with your method. Adam Actually there is a lot wrong with his method if he wants to measure strength or something similar between programs. I also used a full2ply set once, but only for exp...
by Guenther
Sun Oct 30, 2011 10:40 am
Forum: Computer Chess Club: General Topics
Topic: Some large Polyglot books available
Replies: 19
Views: 2632

Re: Some large Polyglot books available

[quote="Guenther]Michel quote: ========1.4.67b================= - Disable "book learning". Polyglot saves learning information but does not use it itself, probably for a reason. I believe the booklearning concept of Polyglot is fundamentally broken. Learning information is engine specific so it sho...
by Guenther
Sun Oct 30, 2011 10:34 am
Forum: Computer Chess Club: General Topics
Topic: Some large Polyglot books available
Replies: 19
Views: 2632

Re: Some large Polyglot books available

My only issue with Polyglot is there is no learning function for books. I have found many times that engines which use Polyglot as the adapter lose the same games over and over again due to lack of learning. So for the time being, I think I will have to try making my books in Arena or something els...
by Guenther
Tue Oct 11, 2011 6:00 am
Forum: Computer Chess Club: General Topics
Topic: Interesting blog post about Komodo...
Replies: 12
Views: 2851

Re: Interesting blog post about Komodo...

If we would listen to random blog news the world would have gone for long...
by Guenther
Thu Aug 25, 2011 12:21 pm
Forum: Computer Chess Club: General Topics
Topic: A new UCI engine ???
Replies: 23
Views: 6582

Re: Maybe Miky .......................

I don't think this 'QN' can be QueeNy (although I posted that as QN.exe), because QueeNy is WinBoard, not UCI... Hi Mr.H.G.Muller ! Take a look - please - to this image from Miky's masterpiece: http://i1238.photobucket.com/albums/ff487/Silvian3/Clipboard02-1.jpg ??? SilvianR It is really QueeNy 0.1...