Search found 647 matches

by Joerg Oster
Fri May 10, 2019 9:08 pm
Forum: Computer Chess Club: General Topics
Topic: Nd4 draws, all other moves lose
Replies: 1
Views: 125

Re: Nd4 draws, all other moves lose

An experimental version of a SF-Mod, at a very early stage, trying to mimic a MCTS-like search, with 3 Threads, 1 GB Hash, 6p syzygy's: info depth 30 seldepth 34 multipv 1 score cp 0 nodes 345553391 nps 3067223 hashfull 976 tbhits 3201076 time 112660 pv e2d4 f1d2 c4b5 e3d4 d5d6 d1a4 b5a4 d2c4 a4b5 c...
by Joerg Oster
Thu May 09, 2019 8:09 am
Forum: Computer Chess Club: General Topics
Topic: A tough nut to crack
Replies: 5
Views: 403

Re: A tough nut to crack

I changed MAX_PLY to 256 in types.h. Rest is latest Stockfish. This change puts a strain on the stack I suppose but it compiles fine and no crash in bench. With your machine it probably is easier to check the whole solution backwards with Joerg's idea and a lot of hash. Maybe counts as a simplifica...
by Joerg Oster
Wed May 08, 2019 8:08 pm
Forum: Computer Chess Club: General Topics
Topic: A tough nut to crack
Replies: 5
Views: 403

Re: A tough nut to crack

To fully resolve this mate in 89 moves would require to search 177 plies deep at minimum!
Stockfish's maximum search depth is 127 plies.

You have to find an engine which can search deep enough.
by Joerg Oster
Fri May 03, 2019 1:43 pm
Forum: Computer Chess Club: General Topics
Topic: interesting position from CCC 8
Replies: 6
Views: 263

Re: interesting position from CCC 8

Qxg5 here (from Komodo MC-Xiphos) seems quite difficult for non-NN engines. It is not obvious that giving up the queen is the most powerful move. 2b3r1/1p1p2pk/2nBp2p/2P1Ppq1/R7/4QBPP/5P2/5K2 w - - Stockfish 10, 4 cores, 30 minutes: depth 54 seldepth 81 multipv 1 score cp 358 nodes 16305499409 nps ...
by Joerg Oster
Wed Apr 17, 2019 4:26 pm
Forum: Computer Chess Club: General Topics
Topic: The problem with adjudication based on score
Replies: 26
Views: 3912

Re: The problem with adjudication based on score

Do any engines correctly score 4k3/rr6/2p1p1p1/1pPpPpPp/pP1P1P1P/P7/8/3K4 w - - 0 1 to be a draw? Yes. :D Just recently added a 1st blockade detection rule to my SF fork. Patch is here: https://github.com/joergoster/Stockfish/commit/9c67b6a6874f2e8f2fb64f94504615020293f1ef position fen 4k3/rr6/2p1p...
by Joerg Oster
Wed Apr 17, 2019 10:27 am
Forum: Computer Chess Club: General Topics
Topic: The problem with adjudication based on score
Replies: 26
Views: 3912

Re: The problem with adjudication based on score

konsolas wrote:
Wed Apr 17, 2019 10:03 am
Nice! Have you considered PR'ing that change to the main version?
No.
It only fixes a corner-case without any measurable benefit.
by Joerg Oster
Wed Apr 17, 2019 10:01 am
Forum: Computer Chess Club: General Topics
Topic: The problem with adjudication based on score
Replies: 26
Views: 3912

Re: The problem with adjudication based on score

Do any engines correctly score 4k3/rr6/2p1p1p1/1pPpPpPp/pP1P1P1P/P7/8/3K4 w - - 0 1 to be a draw? Yes. :D Just recently added a 1st blockade detection rule to my SF fork. Patch is here: https://github.com/joergoster/Stockfish/commit/9c67b6a6874f2e8f2fb64f94504615020293f1ef position fen 4k3/rr6/2p1p...
by Joerg Oster
Sat Apr 06, 2019 8:21 am
Forum: Computer Chess Club: General Topics
Topic: Strange artifact of Syzygy format
Replies: 5
Views: 684

Re: Strange artifact of Syzygy format

Dann Corbit wrote:
Fri Apr 05, 2019 6:22 pm
That's not a mate score.
It shows a win.
By clicking through the mating sequence at https://syzygy-tables.info/?fen=1K6/1R6 ... _w_-_-_0_1
you'll notice many 'only' moves.
Search alone will never find the mating line but by sheer luck.

That's why we use tablebases.
by Joerg Oster
Fri Feb 22, 2019 11:08 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Utility to calculate average depth from PGN files
Replies: 6
Views: 560

Re: Utility to calculate average depth from PGN files

Title says it all.. anyone know of a utility which can parse a PGN file and tell me the average depth searched of program A vs program B? SOMU - function Match stats http://rebel13.nl/download/utilities.html Examples: http://rebel13.nl/text/match-stat-1.htm http://rebel13.nl/text/match-stat-2.htm V...
by Joerg Oster
Fri Feb 22, 2019 10:38 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Utility to calculate average depth from PGN files
Replies: 6
Views: 560

Re: Utility to calculate average depth from PGN files

Title says it all.. anyone know of a utility which can parse a PGN file and tell me the average depth searched of program A vs program B? SOMU - function Match stats http://rebel13.nl/download/utilities.html Examples: http://rebel13.nl/text/match-stat-1.htm http://rebel13.nl/text/match-stat-2.htm V...