Search found 150 matches

by Volker Annuss
Sun Jan 08, 2017 9:17 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Help request for debugging ICS
Replies: 52
Views: 11503

Re: Help request for debugging ICS

I should add that valgrind gives stack traces for many situations that can cause crashs such as using memory that has not been allocaced or already been freed or branching dependent on using uninitialized memory.
by Volker Annuss
Sat Jan 07, 2017 9:46 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Help request for debugging ICS
Replies: 52
Views: 11503

Re: Help request for debugging ICS

Try to compile the ICS with -g and run it under valgrind with valgrind --log-file=ics.log <your ICS start command> and you'll get your stacktrace and more. See the valgrind manual for more options. Expect the ICS to run about 30 times slower than normal. You can download valgrind from http://valgrin...
by Volker Annuss
Thu Jan 05, 2017 7:12 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Another way of evaluating material imbalances
Replies: 4
Views: 1726

Re: Another way of evaluating material imbalances

How do you find efficient rules? Given the rules, how do you estimate the parameters? I did a kind of reverse tuning. I took some fixed values in a fixed order and have an optimizer choose the rules that match them best. In more details, I made a list of 64 rules (better pairs of rules with colours...
by Volker Annuss
Wed Jan 04, 2017 6:15 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Another way of evaluating material imbalances
Replies: 4
Views: 1726

Re: Another way of evaluating material imbalances

You did not misread it. But I suggest to interpret it in a different way. A rule fires when board.material >= rule.material and board.imbalance >= rule.imbalance for every piecetype i. A rule may fire more than once. Every time a rule fires, it contributes to the evaluation and the imbalance is take...
by Volker Annuss
Sun Jan 01, 2017 6:27 pm
Forum: Computer Chess Club: General Topics
Topic: Arminius 2017-01-01 released
Replies: 19
Views: 5270

Re: Arminius 2017-01-01 released

Linux version doesn't run on my Intel i5 (popcount) box. Maybe it's compiled on new hardware(AVX or BMI2) with 'native' flag. When I start from command line I get : Illegal instruction (core dumped) The linux version was compiled with AVX. Now I replaced the archive with a new one that contains Arm...
by Volker Annuss
Sun Jan 01, 2017 5:51 pm
Forum: Computer Chess Club: General Topics
Topic: Arminius 2017-01-01 released
Replies: 19
Views: 5270

Re: Arminius 2017-01-01 released

Popcount version is not running on my system, i7 7500U. Normal 64 bit version runs ok. I downloaded the .zip archive on an i5 6200U system with Windows 10. The popcount version runs ok there. The other x64 version (made for core 2 duo and core 2 quad processors) was blocked by avast, sent to them f...
by Volker Annuss
Sun Jan 01, 2017 3:29 pm
Forum: Computer Chess Club: General Topics
Topic: Arminius 2017-01-01 released
Replies: 19
Views: 5270

Arminius 2017-01-01 released

Arminius 2017-01-01 is out. http://www.nnuss.de/Hermann/Arminius2017-01-01.zip contains Arminius2017-01-01-32Bit.exe (not tested) Arminius2017-01-01-x64.exe (not tested) Arminius2017-01-01-x64-popcnt.exe (fastest, tested with Windows 7, Core i7 860) http://www.nnuss.de/Hermann/Arminius2017-01-01Linu...
by Volker Annuss
Sun Jan 01, 2017 1:07 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Another way of evaluating material imbalances
Replies: 4
Views: 1726

Another way of evaluating material imbalances

Material can be encoded in a single 64 bit number with a four bit counter for every piece type separated by a bit which is always 0. X is used as a special piece type that stands for having the bishop pair. (A single bit would be sufficient.) 00000QQQQ0RRRR0BBBB0NNNN0PPPP0XXXX0qqqq0rrrr0bbbb0nnnn0pp...
by Volker Annuss
Mon Dec 19, 2016 5:21 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Absolute ELO scale
Replies: 58
Views: 14106

Re: Absolute ELO scale

Yesterday I implemented what I see as a natural improvement to a random mover. It is an engine that behaves like a random mover with one exception. Whenever there is a mate in 1 it will mate. Here the results (W-L-D) from 100000 games random mover + mate vs random mover played with cutechess last ni...
by Volker Annuss
Sat Nov 19, 2016 8:00 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: On-line engine blitz tourney November
Replies: 88
Views: 18054

Re: pawn hash

However, the original question has not been answered. What is the best memory allocation ratio for hash to pawn hash? When running multiple tests, a fixed amount of memory is usually available for each engine - for example 256M. Hash is more important than pawn hash. Would a ratio of 230M hash + 26...