Search found 3177 matches

by Sven
Mon Dec 10, 2018 11:34 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: fast(er) movegen
Replies: 24
Views: 602

Re: fast(er) movegen

The move generator of stockfish is ~14x faster than the one of embla (compared via perft 6). So I'm thinking there's some room for improvement in Embla. Currently Embla implements an 8x8 array of pointers. Those pointers point to the 2 x 16 piece-objects. Those objects calculate the moves for their...
by Sven
Mon Dec 10, 2018 11:06 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: UCI2WB 4.0
Replies: 12
Views: 390

Re: UCI2WB 4.0

I think even a binary-only distribution of software whose source code is put under GPL needs to include the correct license information. Why? What would be the legal consequence if it doesn't? IMHO it would mean that whoever downloads it cannot distribute it, or distribute any modified version of i...
by Sven
Sun Dec 09, 2018 8:28 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: UCI2WB 4.0
Replies: 12
Views: 390

Re: UCI2WB 4.0

May it be suggested to include a "readme" file in the executable zip folder for: (1) a summary description, (2) a short start-up example or "usage <>" (3) GNU license, build date and authorship All that is in the source repository. The zip file is just a service to Windows users that are not able t...
by Sven
Fri Dec 07, 2018 9:47 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: For perft nps, what is a node?
Replies: 8
Views: 329

Re: For perft nps, what is a node?

Hi all, In making adjustements to Axolotl, I am unsure about how to calculate nps. 1. During perft (so no evaluation or anything), do you include the last level in the numbers you use to calculate perft? Do you do this even if you do not make and unmake a move at the final level? 2. During regular ...
by Sven
Sun Nov 25, 2018 1:39 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: alpha bound in hash table
Replies: 14
Views: 882

Re: alpha bound in hash table

I think that you should better 1) return STOPSCORE immediately if stopFlag is set after returning from recursive pvs() call, 2) otherwise always store something in the TT, with bestmove possibly being INVALIDMOVE, and 3) remove the "/// ????? if (bestMove == INVALIDMOVE) bestMove = moves[0];" line c...
by Sven
Sun Nov 18, 2018 10:06 am
Forum: Computer Chess Club: Tournaments and Matches
Topic: Open Swiss Edition II
Replies: 58
Views: 2818

Re: Open Swiss Edition II

Please include Jumbo 0.6.66 .
by Sven
Thu Nov 15, 2018 11:55 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: On-line engine blitz tourney November
Replies: 15
Views: 1046

Re: On-line engine blitz tourney November

Jumbo will be on holidays this time.
by Sven
Wed Nov 14, 2018 10:28 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: new Engine, Axolotl
Replies: 31
Views: 2145

Re: new Engine, Axolotl

Hi Guenther, Thanks for your interest! Well spotted, I checked the source code and my engine gets confused when it is about to lose... Seeing as this will happen quite often, I have updated the code in order to cope with this situation. I have done my best to test similar situations to yours, but p...
by Sven
Wed Nov 14, 2018 7:42 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Staged move generation and killers
Replies: 5
Views: 607

Re: Staged move generation and killers

I am currently trying a staged move generator in Minic and want this work flow : * return TT move * Gen Cap and sort them (mvv-lva + see) * return Good Cap * return killers * Gen quiet and sort them (history + pst) * return quiet But how do you ensure, without having generate the quiet moves first,...
by Sven
Sun Nov 11, 2018 12:36 am
Forum: Computer Chess Club: General Topics
Topic: Jumbo 0.6.66
Replies: 0
Views: 268

Jumbo 0.6.66

http://www.schuelebln.de/chess/jumbo

fixes problems that were reported in this thread:

- additional binaries for pre-SSE4.2 CPUs
- show thinking output after each iteration starting with depth 8