Search found 169 matches

by cetormenter
Thu Aug 25, 2016 2:02 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Value of IID
Replies: 6
Views: 1710

Value of IID

Recently I have been playing around with IID in Nirvana and have noticed that it isn't as valuable as it once was. Seeing this I decided to collect some statistics and noticed that in almost 95 percent of cases no pv move is even collected from this search! I figured out that if I were to add an ext...
by cetormenter
Sun Jul 31, 2016 10:03 pm
Forum: Computer Chess Club: General Topics
Topic: Nirvanachess 2.3 Release
Replies: 9
Views: 3053

Nirvanachess 2.3 Release

Hello all,

Nirvanachess 2.3 has been released.

http://nirvanachess.blogspot.com/2016/0 ... lease.html

Overall strength increase is somewhere in the range of 30 elo.
by cetormenter
Mon Jun 13, 2016 4:56 pm
Forum: Computer Chess Club: Tournaments and Matches
Topic: How much stronger will sf and komo be stage 2/3 TCEC?
Replies: 16
Views: 2693

Re: How much stronger will sf and komo be stage 2/3 TCEC?

Nirvana did great this stage. How stronger he is since last TCEC he played ? Around 40 elo or so. The biggest change that increased performance was fixing the bug that caused crashes during the last TCEC. Being on equal footing (20 cores vs 20 cores instead of 8 vs 20) makes a big difference. 10 el...
by cetormenter
Mon Jun 13, 2016 2:16 pm
Forum: Computer Chess Club: Tournaments and Matches
Topic: How much stronger will sf and komo be stage 2/3 TCEC?
Replies: 16
Views: 2693

Re: How much stronger will sf and komo be stage 2/3 TCEC?

The true answer is not strong enough to be noticeable with the number of games that are played in stage 2. Each engine will be playing a total of 30 games which has an error bar of about 90 elo. So unless your engine has improved (or gotten worse by) about 90 elo you would not be able to tell whethe...
by cetormenter
Sun Apr 24, 2016 4:28 pm
Forum: Computer Chess Club: Tournaments and Matches
Topic: CENT - (C)omputer (EN)dgame (T)ournament
Replies: 20
Views: 3794

Re: CENT - (C)omputer (EN)dgame (T)ournament

Will be interesting to see how this turns out. I expect the ratings to be just a compression of what you see on cegt/ccrl since there is less time for the top engines to outplay the weaker ones. I do have one request though. If you could link to my official site instead of a site that I have no affi...
by cetormenter
Sat Apr 16, 2016 3:06 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: On-line engine blitz tourney April
Replies: 12
Views: 2095

Re: On-line engine blitz tourney April

I will be away from my computer when this tournament starts. However I am logged in if you could auto join me that would be great. The version of Nirvana that is playing will be the same one that will be playing TCEC so hopefully it performs well!
by cetormenter
Sun Mar 13, 2016 4:18 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Computer chess progress over say the last 20 years?
Replies: 31
Views: 7037

Re: Computer chess progress over say the last 20 years?

Dear all, I would like to produce a graph showing the progress of the top chess computer over time - in one version as absolute numbers (elo vs year), and in another version, pure algorithmic progress (elo vs year of release, on identical hardware). I can find a lot of information on the web - for ...
by cetormenter
Fri Jan 22, 2016 3:33 pm
Forum: Computer Chess Club: Tournaments and Matches
Topic: The Return 4CPU
Replies: 17
Views: 2893

Re: Standings after Round 18 of 44

The spread is quite small only Alfil and Deep Saros have separated themselves from the pack and even then not by much.
by cetormenter
Sun Jan 03, 2016 4:23 pm
Forum: Computer Chess Club: General Topics
Topic: Endgame positions Stockfish can improve on
Replies: 11
Views: 2038

Re: Endgame positions Stockfish can improve on

Perhaps it was just a bit of bad luck on StockFish's part? Even Nirvana has no issues with finding Bd4 in less than 30 seconds using 4 threads. info depth 30 seldepth 45 multipv 1 score cp 123 nodes 113245946 nps 8507489 time 13310 pv g7f8 d6e5 d3e3 e5f6 e3f3 h2c7 f8h6 e6e5 info depth 31 seldepth 48...