Wednesday, January 31, 2018

Tuesday, January 30, 2018

Laser 1.5 wins 3 League JCER, edition 01/2018 ( Temp - 60').


Laser 1.5 1.6 wins 3 League JCER, edition 01/2018 ( Temp - 60'). Congratulations for: Jeffrey An and Michael An!

Promoted to 2 League: Laser and Senpai.
Demoted to 4 League: ChessBrainVb and Alfil.

P
Engine Jcer Score 1 2 3 4 5 6 7 8 9 0 Chg
1: Laser 1.5 2853 11.5 / 18 XX 01 1= == 01 == == 1= =1 11 +23
2: Senpai 2.0 2919 10.5 / 18 10 XX 01 == =0 01 =1 11 1= == -5
3: Vajolet2 2.5 2838 10.5 / 18 0= 10 XX =1 == == == 10 1= 11 +18
4: Ice3 658 2869 10.0 / 18 == == =0 XX 1= =1 =1 == == == +5
5: SmarThink 1.97 2818 9.0 / 18 10 =1 == 0= XX == =0 01 1= == +9
6: Pedone 1.7 2841 8.5 / 18 == 10 == =0 == XX 11 =0 =0 == -2
7: Kohai 1.0 2844 8.5 / 18 == =0 == =0 =1 00 XX =1 01 =1 -4
8: Wasp 2.60 2820 8.5 / 18 0= 00 01 == 10 =1 =0 XX 11 == +4
9: ChessBrainVB 3.61 2829 7.0 / 18 =0 0= 0= == 0= =1 10 00 XX =1 -13
10Alfil 15.8 2871 6.0 / 18 00 == 00 == == == =0 == =0 XX -38

Temp 60' Hardware: Intel Core i7-4710MQ with 7,9 GB Memory, 
Windows 10 64-bit, 90 Tournament games
GUI-Arena 3.5.1, Book: Perfect 2017Games, engines, book - download    Table created - Scid vs PC




Stockfish 18012814_Rocky640 - new version!

Stockfish, chess engine UCI

Rating JCER = 3389
🔬 Author: syzygy1

Timestamp: 1517147805 

Limit the king distance factor when evaluating passed pawns (#1373) 

Limit the king distance factor when evaluating passed pawns 
Passed STC 

LLR: 3.31 (-2.94,2.94) [0.00,5.00] 
Total: 23987 W: 5550 L: 5281 D: 13156 
and LTC 

LLR: 2.97 (-2.94,2.94) [0.00,5.00] 
Total: 16926 W: 3014 L: 2820 D: 11092 

Bench: 5059457

⭐⭐⭐   Stockfish 18012814_Rocky640  - download

Monday, January 29, 2018

Chess Engine: BrainFish 180128 and book: Cerebellum Light 180128

BrainFish - UCI chess engine
JCER Rating=3378

Information from the author:
Brainfish is a standard Stockfish chess engine extended by a general book format that is capable of handling a reduced part of Cerebellum, which is an innovative chess opening and playing book. In Brainfish, the book moves are only used in engine games, not in analysis mode.

All moves in the published book Cerebellum_light.bin generated by Cerebellum were calculated by using
Stockfish as analysis engine, without using statistics. Then those moves where recalulated by the Cerebellum Library using a graph algorithm that makes all scores in the library consistant. Consistant scores means for example, in case the starting position has a score of 0.2 and a best move is 1.e4, the position after 1.e4 must have the score -0.2. Additionally, some others properties are found, like the best main line of a position and possible Transitions.

Hence, when Brainfish is playing moves that are in the Cerebellum book, it plays like Stockfish regarding the evaluation, only at a much higher skill level. Pondering is only active after the last move out of the book has been played.

The lines in Cerebellum_light.bin are in some cases reaching the late middlegame and endgame.
BrainFish is licensed unter the GPL, which includes the access code to the general book format.

Cerebellum Light 180128 - download   BrainFish 180128 - download


Nemorino 4.5 - UCI chess engine. New version!

Nemorino - UCI chess engine
Author - Christian Günther

Rating JCER=2884

Nemorino 4.5 download

Chess22k 1.6 wins 6 League JCER, edition 01/2018 ( Temp - 60').


Chess22k 1.6 wins 6 League JCER, edition 01/2018 ( Temp - 60'). Congratulations for: Evgeniy Korniloff!

Promoted to 5 League: Chess22k and Sctr.
Demoted to 7 League: Ruydos and ProDeo.


P
Engine Jcer Score 1 2 3 4 5 6 7 8 9 0 Chg
1: Chess22k 1.6 2602 12.5 / 18 XX 10 0= 11 1= == 1= 11 10 11 +41
2: Sctr 1.1f 2858 11.5 / 18 01 XX 10 == 00 1= 1= =1 11 11 -32
3: Tornado 8 2644 11.0 / 18 1= 01 XX =0 =1 10 == 11 =0 11 +14
4: Murka 3 2637 10.5 / 18 00 == =1 XX =1 =0 10 == 11 11 +11
5: Gnucheese 1.00 2602 10.0 / 18 0= 11 =0 =0 XX 0= 01 11 1= 1= +18
6: Baron 3.42 2603 9.5 / 18 == 0= 01 =1 1= XX =1 01 =0 01 +13
7: Scorpio 2.78 2604 9.5 / 18 0= 0= == 01 10 =0 XX == 11 11 +13
8: Rhetoric 1.4.3 2567 7.0 / 18 00 =0 00 == 00 10 == XX 1= 11 -2
9: Ruydos 1.0.27 2572 6.5 / 18 01 00 =1 00 0= =1 00 0= XX =1 -9
10: ProDeo 2.6 2577 2.0 / 18 00 00 00 00 0= 10 00 00 =0 XX -56

Temp 60' Hardware: Intel Core i7-4710MQ with 7,9 GB Memory, 
Windows 10 64-bit, 90 Tournament games
GUI-Arena 3.5.1, Book: Perfect 2017Games, engines, book - download    Table created - Scid vs PC





Sunday, January 28, 2018

Stockfish 18012814 - new version!

Stockfish, chess engine UCI

Rating JCER = 3389
🔬 Author: syzygy1

Timestamp: 1517146807 

This modifies the in-search TB probing to continue searching for a mate "behind" a TB win (or loss). (#1285) 

It does the following: 
- If a TB win or loss value allows an alpha or beta cutoff, the cutoff is taken. 
- Otherwise, the search of the current subtree continues. In PV nodes, the final value returned is adjusted to reflect that the position is a TB win (or loss). 

The patch also fixes a potential problem caused by root_probe() and root_probe_wdl() dirtying the root-move scores. 

This patch removes the limitation of current master that a mate is never found if the root position is not yet in the TBs, but the path to mate at some point enters the TBs. The patch is intended to preserve the efficiency and effectiveness of the current TB probing approach. 

No functional change (withouth TB)
⭐⭐⭐   Stockfish 18012814 - download

Only for donors added: 56 chess engines and 2 book

Only for donors added: 56 chess engines and 2 book:

andscacs092165, andscacs093, Arasan 20.4, Arasan 20.4.1, asmfish-012018+mf, atlas 3.90, atlas 3.91, baron 3.42, BrainFish 180106, BrainFish 180113, BrainFish 180124, BugChess 2.41, ChessV 2.1, Clever-girl-rev160, Dimitri 393, Dorpsgek-Eves-Temptation, Ethereal 8.80, galjoen v0.37.2, greko 2017, Hactar 0.9.0, OpenTal 1.1, Pedone 1.7, Popochin 4.0, Raubfisch ME262_GTZ16b, Raubfisch ME262_GTZ16b1, Raubfisch ME262_GTZ16b3, Raubfisch X37b, Raubfisch X37b1, Raubfisch X37b3, RomiChessP3nOK, RubiChess-0.7-graham, Schooner 1.8, Shield 2.1, Skiull 0.3, snowy 0.2, Soberango 0.12.0, sting-sf-95, sting-sf-96, stockfish 18010113, stockfish 18010113b, stockfish 18010422, stockfish 18011308, stockfish 18011308_Ilvec, stockfish 18011309, stockfish 18011309_Costalba, stockfish 18012314, stockfish 18012314_Nicolet, stockfish 18012314_Vijlbrief, stockfish 18012416, SugaR.XPrO 220118, texel108a8, TogaII401, Vajolet2 2.5, waxman 2017, Zevra v1.7.1 r563, Zevra v1.8 r583.

Book: Cerebellum_light 180103, Cerebellum_light 180109.



ChessV 2.1 - universal chess program with a graphical user interface!

ChessV 2.1 - download



ChessV is a free, open-source universal chess program with a graphical user interface, sophisticated AI engine, and other features of traditional Chess programs. 
As a "universal" chess program, it not only plays orthodox Chess, it is also capable of playing games reasonbly similar to Chess. 





6 League JCER, edition 01/2018 P-60' - after round 12


REngineAuthorJcerScore1234567890S-B
01Chess22k 1.6 Evgeniy Korniloff26029.5/12· ·0111===11111 51,25 
02Tornado 8 Engin Üstün26448.0/121· ·=0=111==011 40,00 
03Murka 3  26378.0/120=· ·===1==1111 38,75 
04Sctr 1.1f Can Cetin & Dogac Eldenk28588.0/1201==· ·01=1=111 38,25 
05Gnucheese 1.00 26026.5/120===01· ·010111 35,25 
06Baron 3.42 Richard Pijl26035.5/12==0=01· ·0=1=01 30,25 
07Rhetoric 1.4.3Alberto Sanjuan25675.5/120000==01· ·=111 21,50 
08Scorpio 2.78 Daniel Shawul26045.0/120=00=10=0=· ·11 24,00 
09Ruydos 1.0.27 Alvaro Begue & Jose Manuel Moran25722.5/120=100000=00· ·= 15,50 
10ProDeo 2.6 Ed Schröder25771.5/1200000010000=· · 6,75 

Atlas 3.91 - UCI chess engine. New version (bug fixed)!

Atlas - UCI chess engines
Author: Andreas Manzanares 
Rating JCER=2661

Atlas 3.91 - download

All chess engines downloadable on the site


Saturday, January 27, 2018

Chess engine: Zevra 1.8 build 583

Zevra - Rating JCER=1761
Author - Oleg Smirnov 

 Zevra 1.8 b583 - download

Chess engine: Popochin 4.0

Popochin - winboard chess engine
Author: Miguel Izquierdo

Popochin 4.0 - download


3 League JCER, edition 01/2018 P-60' - after round 9



REngineAuthorJCERScoreVaPeIcLaSeSmWaKoAlChS-B
01Vajolet2 2.5 Belli Marco28386.0/9·==01=1=11 24,75 
02Pedone 1.7 Fabio Gobbato28415.5/9=·==1==1== 24,00 
03Ice3 658 Thomas Petzke28695.0/9==·==1==== 22,25 
04Laser 1.5 Jeffrey An and Michael An28535.0/91==·001=1= 22,00 
05Senpai 2.0 Fabien Letouzey29195.0/900=1·=1==1 20,25 
06SmarThink 1.97 Sergei Markoff28184.5/9==01=·0==1 19,75 
07Wasp 2.60 John Stanback28204.0/90==001·==1 16,25 
08Kohai 1.0 Fabien Letouzey, Michael Byrne28443.5/9=0=====·=0 16,50 
09Alfil 15.8 Enrique Sánchez Acosta28713.5/90==0====·= 15,25 
10ChessBrainVB 3.61Evgeniy Korniloff28293.0/90===0001=· 13,00 

Thursday, January 25, 2018

Chess engine: Raubfisch ME262 GTZ16b3

Raubfisch is an engine derived from Stockfish and Torpedo 1.1 XJR (private engine by Frank Karl Werner and Max Fehler)

Rating JCER - 3379

Raubfisch ME262 GTZ16b3 - download

Stockfish 18012416 - new version!

Stockfish, chess engine UCI

Rating JCER = 3389
🔬 Author: Mibere

Timestamp: 1516809244 

Top CPU Contributors as of January 2018 (#1367) 

No functional change.
⭐⭐⭐   Stockfish 18012416 - download

Chess engine: Raubfisch X37b3

Raubfisch is an engine derived from Stockfish and Torpedo 1.1 XJR (private engine by Frank Karl Werner and Max Fehler)

Rating JCER - 3379

Raubfisch X37b3 - download

Chess Engine: BrainFish 180113 and book: Cerebellum Light 180113

BrainFish - UCI chess engine
JCER Rating=3378

Information from the author:
Brainfish is a standard Stockfish chess engine extended by a general book format that is capable of handling a reduced part of Cerebellum, which is an innovative chess opening and playing book. In Brainfish, the book moves are only used in engine games, not in analysis mode.

All moves in the published book Cerebellum_light.bin generated by Cerebellum were calculated by using
Stockfish as analysis engine, without using statistics. Then those moves where recalulated by the Cerebellum Library using a graph algorithm that makes all scores in the library consistant. Consistant scores means for example, in case the starting position has a score of 0.2 and a best move is 1.e4, the position after 1.e4 must have the score -0.2. Additionally, some others properties are found, like the best main line of a position and possible Transitions.

Hence, when Brainfish is playing moves that are in the Cerebellum book, it plays like Stockfish regarding the evaluation, only at a much higher skill level. Pondering is only active after the last move out of the book has been played.

The lines in Cerebellum_light.bin are in some cases reaching the late middlegame and endgame.
BrainFish is licensed unter the GPL, which includes the access code to the general book format.

Cerebellum Light 180113 - download   BrainFish 180124 - download



Stockfish 18012314_Nicolet - new version!

Stockfish, chess engine UCI

Rating JCER = 3389
🔬 Author: Stéphane Nicolet

Timestamp: 1516714005 
Contempt 20 
Set the default contempt value of Stockfish to 20 centipawns. 
The contempt feature of Stockfish tries to prevent the engine from simplifying the position too quickly when it feels that it is very slightly behind, instead keeping the tension a little bit longer. 

Various tests in November 2017 have proved that our current imple- mentation works well against SF7 (which is about 130 Elo weaker than current master) and than the Elo gain is an increasing function of contempt, going (against SF7) from +0 Elo when contempt is set at zero centipawns, to +30 Elo when contempt is 40 centipawns. 

This november discussion left open the decision of which "default" value for contempt we should use for Stockfish, taking into account the various uses ofStockfish (opening preparation for humans, computer online tournaments,analysis tool for web pages, human/computer play, etc). 

This pull request proposes to set the default contempt value of SF to twenty centipawns, which turns out to be the highest value which is not a regression against current master, as this seemed to be a good compromise between risk and safety. A couple of SPRT[-3..1] tests were done to bisect this value: 
Contempt 10: http://tests.stockfishchess.org/tests/view/5a5d42d20ebc5902977e2901 (PASSED) 
Contempt 15: http://tests.stockfishchess.org/tests/view/5a5d41740ebc5902977e28fa (PASSED) 
Contempt 20: http://tests.stockfishchess.org/tests/view/5a5d42060ebc5902977e28fc (PASSED) 
Contempt 25: http://tests.stockfishchess.org/tests/view/5a5d433f0ebc5902977e2904 (FAILED) 

Surprisingly, a test at "very long time control" hinted that using contempt 20 is not only be non-regressive against contempt 0, but may actually exhibit some small Elo gain, giving a likehood of superio-rity of 88.7% after 8500 games: 

VLTC: 
ELO: 2.28 +-3.7 (95%) LOS: 88.7% 
Total: 8521 W: 1096 L: 1040 D: 6385 

Finally, there was some concerns that a contempt value of 20 would be worse than a value of 7, but a test with 20000 games at STC was neutral: 

STC: 
ELO: 0.45 +-3.1 (95%) LOS: 61.2% 
Total: 20000 W: 4222 L: 4196 D: 11582 

See the comments in pull request 1361 for the long, nice discussion (180 entries :-)) leading to the decision to propose contempt 20 as the default value: 
https://github.com/official-stockfish/Stockfish/pull/1361 

Whether Stockfish should strictly adhere to the Komodo and Houdini semantics and add the UCI commands to force the contempt to be White in the so-called "analysis mode" is still under discussion, and may be or may not be the object of a future commit. 

Bench: 5783344 

⭐⭐⭐   Stockfish 18012314_Nicolet - download


6 League JCER, edition 01/2018 P-60' - after round 6



REngineAuthorJCERScoreScChToMuGnScRhBaRuPrS-B
01Sctr 1.1f Can Cetin & Dogac Eldenk28584.5/6·1=0111 12,00 
02Chess22k 1.6 Evgeniy Korniloff26024.5/6·0111=1 11,00 
03Tornado 8 Engin Üstün26444.0/601·=1=1 10,50 
04Murka 3 26374.0/6=·===11 8,25 
05Gnucheese 1.0026023.0/610==·01 9,50 
06Scorpio 2.78 Daniel Shawul26043.0/6001·==1 6,50 
07Rhetoric 1.4.3Alberto Sanjuan25673.0/600==·11 5,50 
08Baron 3.42 Richard Pijl26032.0/60===·=0 6,25 
09Ruydos 1.0.27 Alvaro Begue & Jose Manuel Moran25721.0/60=000=· 3,00 
10ProDeo 2.6 Ed Schröder25771.0/6000001· 2,00 

Best posts of the month