TalkChess.com

Latest Stockfish Dev (2019-09-16, one thread, own compile, no tablebases). It's a monster :mrgreen: Oh well, at least it is getting there, currently failing high for the eleventh time and searching with a reduced depth of 67 plies. Ten hours and counting. rk6/p1r3p1/P3B1Kp/1p2B3/8/8/8/8 w - - Engine...

Probably the use of tablebases? There are more winning moves but without tablebases most of the time Stockfish latest version plays 1...g5 Only some Fail Highs for 1..h5 but not played in a full PV 8/p5pp/3k1p2/3p4/1P3P2/P1K5/5P1P/8 b - - Engine: Dev 20190916 (512 MB) by T. Romstad, M. Costalba, J. ...

Don't think there is much going on in the position, not that I can see. Old Ancalagon only gives +0.08 everywhere, this was pre contempt, I believe it had something to do with the side to move bonus but I'm nor sure anymore. Or was it a primitive contempt bonus? But that was the most interesting thi...

Maybe we can look at the position again, with some improved programming in more than a year and a half and some people with very big machines :)? Maybe see new things as requested also here http://www.talkchess.com/forum3/viewtopic.php?f=2&t=69361&start=70#top The questions still pending for these p...

Very preliminary report. Crystal v1.1 (bench 7410759) segfaulted: kernel: [84966.051797] crystal[26855]: segfault at b3a0000017d ip 000000000042095f sp 00007f4342fc6490 error 4 in crystal[400000+3e000] Using 16 threads, 16384 MB hash, all other settings at default. Linux. [EDIT] Was actually using ...

Good to hear! But if you have more details is there maybe something the administrators could learn, why this site was slowed down in your case? I can imagine that not being https would mean some different behaviour, that might be something to look at perhaps..

I unnderstand from your post in the Help and Suggestions subforum Louis, that it was when visiting the website from your University network? No delays here but occasionally when I have not been here for a while, when you go to the main index the site wants to load some new code(?) and there is a red...

Here is another simple BMI2 compile with TDM-GCC. Had to post it in the Rybka forum because this forum won't accept files this large. So thanks to the Rybka forum! A Crystal BMI2 compile Bench: =========================== Total time (ms) : 4665 Nodes searched : 7410759 Nodes/second : 1588587 (Compil...

Kaissa still needs depth 34 to find it, with one thread, 512 Mb. With only 16 Mb it is a bit slower, depth 39. 36/76 1:53 +M8++ 1.Bd7 (274.642.680) 2429 36/76 1:53 +M5 1.Bd7 b4 2.Kf7 b3 3.Ke7 b2 4.Kd8 g5 5.Bxc7+ (274.707.943) 2429 37/10 1:53 +M5 1.Bd7 b4 2.Kf7 b3 3.Ke7 b2 4.Kd8 g5 5.Bxc7+ (275.052.3...

Black still has a plus, but to me looks like White has reasonable chances to draw, in spite of score, after 30... e5 {instead of 30... a5?} 31. a5 exf4 32. gxf4 Kh7 33. f5 {=+} Kh6 34. Rxa6 Kg5 35. Nd4 2r1r3/6p1/R4p2/P4Pk1/2PN2p1/3P4/5K1P/8 b - - Engine: Kaissa IV NoContempt (not the best version, o...