Search found 3989 matches

by syzygy
Sat Mar 02, 2019 4:15 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Testing the implementation of Syzygy
Replies: 1
Views: 510

Re: Testing the implementation of Syzygy

First make sure that probing TB positions returns the correct results (e.g. by comparing with what SF returns when given the 'd' command). Once that works, you could try some of the "Longest phases" positions you can find at https://syzygy-tables.info/ by setting up a position with the material you ...
by syzygy
Sat Mar 02, 2019 4:07 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Training the trainer: how is it done for Stockfish?
Replies: 22
Views: 2938

Re: Training the trainer: how is it done for Stockfish?

Several people are already trying (I am not one of them). As you can appreciate, writing a competitive chess engine is not a trivial exercise (although relatively straightforward these days compared to say 20 years ago). Combining two very different engine architectures is even more difficult. Inci...
by syzygy
Sat Mar 02, 2019 2:23 pm
Forum: Computer Chess Club: General Topics
Topic: 2950x SMT ON and SMT OFF
Replies: 13
Views: 1302

Re: 2950x SMT ON and SMT OFF

Just a quick and dirty example of the speed up of using SMT ON 16 Cores + 32 threads vs. SMT Off + 16 Cores + 16 Threads SMT On: NPS 32702 39.9% better then SMT Off. SMT On: Time to depth 238s 19.3% better then SMT Off Time to depth is the real performance increase, and why you should never use NPS...
by syzygy
Sun Feb 17, 2019 4:16 pm
Forum: Computer Chess Club: General Topics
Topic: End of Era is there: SF is finally beaten!
Replies: 118
Views: 16621

Re: End of Era is there: SF is finally beaten!

Like another user pointed out, "How are we sure we are playing against chess engine and not the games database? And when does the multimillion game database end and the chess engine begin ?" The answer is: by understanding what Leela does. (Or, if one does not have the time or inclination to study ...
by syzygy
Sat Feb 16, 2019 7:27 pm
Forum: Computer Chess Club: General Topics
Topic: End of Era is there: SF is finally beaten!
Replies: 118
Views: 16621

Re: End of Era is there: SF is finally beaten!

... There is no any "spectacular" and there is no any "plan" on the side of Leela. In reality the position after 19.-c4 and the good continuation is in the 32930net of Leela. To prove this I used the weak GTX 1060 to find the 20.Bg6 move. From net30100 to net32000 Leela want to move 20.Be4, 20.Bxc4...
by syzygy
Sun Feb 03, 2019 8:05 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: chess programming - predictions for the next 5 years
Replies: 15
Views: 2512

Re: chess programming - predictions for the next 5 years

let's address the first point, improving tactics... it seems ironic that I not only suggest but RECOMMEND improving tactics... aren't AB searchers tactical beasts? well, not really... :) what happened was that, once engines bypassed humans in tactics, their tactical sharpness started to degrade... ...
by syzygy
Sun Jan 27, 2019 9:21 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Recommended important files to keep on SSD for 7-piece Syzygy
Replies: 3
Views: 772

Re: Recommended important files to keep on SSD for 7-piece Syzygy

Don't the files with pawns in them require the underlying files with the pieces promoted? Not if you don't care too much about being able to win a winning TB position. A probe of a 7-piece WDL table may do probes of 6-piece WDL subtables (reached through captures), which means you need those subtab...
by syzygy
Sun Jan 13, 2019 7:38 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: compiling stockfish with more options
Replies: 11
Views: 1400

Re: compiling stockfish with more options

Besides i already use stockfish derivatives with many options, i just was curious if i would be able to make one of them on my own, but its not most important. So it is possible and not all too difficult, but it requires a bit of knowledge of C++ and several modifications to the source code. Since ...
by syzygy
Sun Jan 13, 2019 1:03 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: compiling stockfish with more options
Replies: 11
Views: 1400

Re: compiling stockfish with more options

Here is my problem. I can add new tuning option just visually by adding something like "o["X"] << Option(100, 0, 100);" in ucioption.cpp but i cant actually connect it in proper way to some value. For example i want for "X" to change parameters in types.h file, we can say "QueenValueMg". QueenValue...
by syzygy
Sun Jan 13, 2019 2:10 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: compiling stockfish with more options
Replies: 11
Views: 1400

Re: compiling stockfish with more options

I have no problem with changing values by this tool, but its quite time-consuming to make a guess and then make .exe one by one with small fluctuations of values. If i have possibility to change values from uci menu in gui it would be incomparably faster for engine tests (faster like faster is copy...