Saturday, September 30, 2017

Stockfish 17092916 - new version!

Stockfish, chess engine UCI

Rating JCER = 3401
🔬 Author: Joost VandeVondele

Timestamp: 1506695952 

Measure nodes after search finished. 
Only affects nmpsec in the multithreaded case. 

No functional change. 

⭐⭐⭐   Stockfish 17092916 - download

Chess engine: Zorgus

Zorgus - UCI/Winboard chess engine.
This chess engine is based on the work of George Georgopoulos and Tom Kerrigan.

Author - Olivier de Lannoy

Zorgus - download

Chess engine: Cfish 300917

Rating JCER=3339

Cfish 300917 - download

Chess engine: Shallow 1.0.3

Shallow 1.0.3 download

Author - Dmitry Sultanov

Rating JCER=2268

Chess engine: Sf290917MZ

Author - Marco Zerbinati

 Sf290917MZ - download

Chess engine: Marvin 2.2.0

Marvin - Winboard/UCI chess engine

Rating JCER=2402


Author - Martin Danielsson

Marvin 2.2.0 - download

Jurek Chess Engines Rating 01.10.2017

All list - pdf file     51.250 games JCER - download

Friday, September 29, 2017

Shark 170827bf wins JCER Test Tournament, 2017.09.21 - 2017.09.29

Temp 3' + 3" Hardware: Intel Core i3-3217U CPU 3,9 GB Memory , Windows 10 x64
924 Tournament games, GUI-Arena 3.5.1161.651 JCER games,  until 31.12.2016
45.406 JCER games (from 01.01.2017)  JCER League
Book: Perfect 2017,   Table created - Scid vs PC

Chess engine: Symphysodon 290917

Symphysodon 290917 - download

Rating JCER=3389

Author - Jose Perez

Chess Engine: BrainFish 170923 and book: Cerebellum Light 170925

BrainFish - UCI chess engine

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 - download   BrainFish 170820 - download

SugaR XPrO 280917 - very strong chess engine!

Sugar - UCI chess engine

Author - Marco Zerbinati 


SugaR XPrO 280917 - download

Chess engine: Rubichess 0.5TM

Rubichess 0.5TM - download

Chess engine: Dimitri 3.81

Dimitri - winboard chess engine
Author - Luigino Viscione

JCER rating=2151

Dimitri 3.81 - download

Chess engine: asmFish 250817-tactic

JCER Rating=3409

asmFish 250817-tactic - download

Stockfish 17092216 - new version!

Stockfish, chess engine UCI

Rating JCER = 3427
🔬 Author: GuardianRM

Timestamp: 1506091688 

Tweak statScore condition 
The first change (ss->statScore >= 0) does nothing. 

The second change ((ss-1)->statScore >= 0 ) has a massive change. (ss-1)->statScore is not set until (ss-1) begins to apply LMR to moves. 
So we now increase the reduction for bad quiets when our opponent is running through the first captures and the hash move. 

LLR: 2.95 (-2.94,2.94) [0.00,4.00] 
Total: 57762 W: 10533 L: 10181 D: 37048 

LLR: 2.95 (-2.94,2.94) [0.00,5.00] 
Total: 19973 W: 2662 L: 2480 D: 14831 

Bench: 5037819 

⭐⭐⭐   Stockfish 17092216 - download

Wednesday, September 27, 2017

Jumbo 0.5.3 - winboard chess engine. New version!

Jumbo - winboard chess engine

Important notes:
The Win64 binary was built using CygWin GCC 5.4.0 and requires the following DLLs that are part of the CygWin installation:

Rating JCER=2199
Author - Sven Schüle, Berlin (Germany)

Jumbo 0.5.3 - download

SugaR XPrO 1.3 - very strong chess engine!

Sugar - UCI chess engine

Author - Marco Zerbinati 


SugaR XPrO 1.3 - download

Saturday, September 23, 2017

Only for donors added: 54 chess engines and book

Only for sponsors added 53 chess engines (folder: ChessEnginesFree):
andscacs0.92.7z Aristides_v2.0d_010917.7z BrainFish_170831.7z bullitchess_1.0.7z cheese-1.91.7z ChessBrainVB_V3_50.7z CorChess_1.8_070917_x32-x64.7z demolito_20170826.7z DirtyBit.0.39.5.t5.7z EnkoChess-010917.7z Ethereal8.27.7z Fire_6.7z galjoen_v0. ISA 2_0_48.7z Joker Klasman Father.7z McBrain2.7.7z Monolith_03.7z Nemorino_3.04_win64.7z Orka 170917.7z Raubfisch ME262_GTZ13b.7z Raubfisch ME262_GTZ15.7z Raubfisch ME262_GTZ15a.7z Raubfisch ME262_GTZ15b..7z Raubfisch X36.7z Raubfisch X36a.7z Raubfisch X36b.7z RuFuS yz-r01.7z RuFuS yz-r02.7z RuFuS yz-r03.7z ruydos sctr_1.0.7z Sf030817MZ.7z Shark 170827 x64 bf.7z Shark 170917 bf.7z stockfish_17082611.7z stockfish_17083015.7z stockfish_17083108.7z stockfish_17090120.7z stockfish_17090120_costalba.7z stockfish_17090120b.7z stockfish_17090209.7z stockfish_17090510.7z stockfish_17090510_syzygy.7z stockfish_17090912.7z stockfish_17091212.7z stockfish_17091709.7z stockfish_17091710.7z texel107a29.7z texel107a35.7z Toltec Sambala.7z zurichess-neuchatel.7z

Book: Mini4C

Free chess engines - download update

G2- 145 chess engines
asmFish - 39 files

Fire 6.0 - new version best UCI chess engines!

Fire - UCI chess engine

author compilation - by Norman Schmidt

Rating JCER=3164

Fire 6.0 download

Chess engine: Monolith 0.3

Monolith - UCI chess engine
Author - Jonas Mayr
Rating JCER=1406

Monolith 0.3 - download

Chess engine: Joker (2) 0.7.7

Joker (2) - UCI chess engine
Author - Manlio Morini

Joker (2) 0.7.7- download

Chess engine: Dirty-Bit 0.39

Dirty-Bit - UCI chess engine
Author - Andrew Backes

Dirty-Bit 0.39 - download

Thursday, September 21, 2017

Raubfisch ME262 GTZ15 wins JCER Test Tournament, 2017.09.14 - 2017.09.18

Temp 3' + 3" Hardware: Intel Core i3-3217U CPU 3,9 GB Memory , Windows 10 x64
480 Tournament games, GUI-Arena 3.5.1161.651 JCER games,  until 31.12.2016
45.406 JCER games (from 01.01.2017)  JCER League
Book: Perfect 2017,   Table created - Scid vs PC

Shark 170917bf (has Cerebellum book integrated in exe file.)

Previous version chess engine Shark 

Rating JCER=3395

Shark 170917bf download

Texel 1.07a35 - UCI Chess engine. New version

Texel - UCI Chess engines

Author - Peter Osterlund

Rating JCER=2979 (19 place in Super League)

Texel 1.07a35 download

Stockfish 17091710 - new version!

Stockfish, chess engine UCI

Rating JCER = 3427
🔬 Author: Syzygy 

Timestamp: 1505637850 

Let ss->ply denote the number of plies from the root to the current node 

This patch lets ss->ply be equal to 0 at the root of the search. 

Currently, the root has ss->ply == 1, which is less intuitive: 

- Setting the rootNode bool has to check (ss-1)->ply == 0. 

- All mate values are off by one: the code seems to assume that mated-in-0 
is -VALUE_MATE, mate-1-in-ply is VALUE_MATE-1, mated-in-2-ply is VALUE_MATE+2, etc. 
But the mate_in() and mated_in() functions are called with ss->ply, which is 1 in 
at the root. 

- The is_draw() function currently needs to explain why it has "ply - 1 > i" instead 
of simply "ply > i". 

- The ss->ply >= MAX_PLY tests in search() and qsearch() already assume that 
ss->ply == 0 at the root. If we start at ss->ply == 1, it would make more sense to 
go up to and including ss->ply == MAX_PLY, so stop at ss->ply > MAX_PLY. See also 
the asserts testing for 0 <= ss->ply && ss->ply < MAX_PLY. 

The reason for ss->ply == 1 at the root is the line "ss->ply = (ss-1)->ply + 1" at 
the start for search() and qsearch(). By replacing this with "(ss+1)->ply = ss->ply + 1" 
we keep ss->ply == 0 at the root. Note that search() already clears killers in (ss+2), 
so there is no danger in accessing ss+1. 

I have NOT changed pv[MAX_PLY + 1] to pv[MAX_PLY + 2] in search() and qsearch(). 
It seems to me that MAX_PLY + 1 is exactly right: 

- MAX_PLY entries for ss->ply running from 0 to MAX_PLY-1, and 1 entry for the 
final MOVE_NONE. 

I have verified that mate scores are reported correctly. (They were already reported 
correctly due to the extra ply being rounded down when converting to moves.) 

The value of seldepth output to the user should probably not change, so I add 1 to it. 
(Humans count from 1, computers from 0.) 

A small optimisation I did not include: instead of setting ss->ply in every invocation 
of search() and qsearch(), it could be set once for all plies at the start of 
Thread::search(). This saves a couple of instructions per node. 

No functional change (unless the search searches a branch MAX_PLY deep), so bench 
does not change. 

⭐⭐⭐   Stockfish 17091710 - download

Orka 170917 (Stockfish clone) - new version

Previous version chess engine Orka 

Rating JCER=3367

Orka 170917 - download

Stockfish 17091710 - new version!

Stockfish, chess engine UCI

Rating JCER = 3427
🔬 Author: Stéphane Nicolet 

Timestamp: 1505634747 
Score unopposed weak pawns only if majors 

Do not use the opposed flag for scoring backward and isolated pawns 
in pawns.cpp, instead give a S(5,25) bonus for each opponent unopposed 
weak pawns when we have a rook or a queen on the board. 

STC run stopped after 113188 games: 
LLR: 1.63 (-2.94,2.94) [0.00,5.00] 
Total: 113188 W: 20804 L: 20251 D: 72133 

LLR: 2.95 (-2.94,2.94) [0.00,5.00] 
Total: 66673 W: 8672 L: 8341 D: 49660 

This is Alain Savard's idea, just with a different bonus. 
Bench: 6259498 

Nemorino 3.04 - UCI chess engine. New version!

Nemorino - UCI chess engine

Author - Christian Günther
Home page

Rating JCER=2901

Nemorino 3.04 download

Wednesday, September 20, 2017

Wednesday, September 13, 2017

Stockfish 17091212 - new version!

Stockfish, chess engine UCI

Rating JCER = 3427
🔬 Author: Ilvec

Timestamp: 1505212313 

Higher Move Overhead 

This shoudl reduce time losses experienced by users after new time management code. 

Verified for no regression in very short TC (4sec + 0.1) 
LLR: 2.95 (-2.94,2.94) [-3.00,1.00] 
Total: 35262 W: 7426 L: 7331 D: 20505 

Bench 5322108 

⭐⭐⭐   Stockfish 17091212 - download

Chess engine: Raubfisch X36b

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

Rating JCER - 3341
Raubfisch X36b - download

Stockfish 17090912 - new version!

Stockfish, chess engine UCI

Rating JCER = 3427
🔬 Author: Ianfab

Timestamp: 1504952614 

Extend ShelterWeakness array by dimension isKingFile 

Use different penalties for weaknesses in the pawn shelter 
depending on whether it is on the king's file or not. 

LLR: 2.96 (-2.94,2.94) [0.00,5.00] 
Total: 71617 W: 13471 L: 13034 D: 45112 

LLR: 2.95 (-2.94,2.94) [0.00,5.00] 
Total: 48708 W: 6463 L: 6187 D: 36058 

Bench: 5322108 

⭐⭐⭐   Stockfish 17090912 - download

Chess engine: SCTR 1.0

SCTR 1.0 - download

Authors: Can Cetin & Dogac Eldenk

Tuesday, September 12, 2017

Chess engine: Klasman Father

Klasman Father - download

Isa 2.0.48 - winboard chess engine (Windows and Linux)

Isa - winboard chess engine

About engine: Isa is a chess engine I developp during my free time. Chess programming is my hobby since the 80's years. I started developing with QuickBasic 4.5 (I wrote a weak engine named Bishop) .
Then I learned the C langage and studed the TSCP source code (like everyone I think) .
« Titan » and « DCP » (DCP = Titan + winboard protocol) was born . Strengh was always so weak , around 1400 elo .
After that I rewrote again my engine and Jars came out (I released Jars) and then Yoda (both played around 1800-1900 elo.
After finding dificult to improve , I stopped working on both engines and started to work on Isa , a diferent way.
I started Isa from Firstchess.c source.

Author - Daniel Anulliero

Rating JCER=2078

Isa 2.0.48 - download

TCEC Season 10 is starting about October 10th

TCEC Season 10

Stage 1 - preliminary stage.
Involving 24 engines, playing a single round robin.
Time control: 60 min. + 10 sec/move.
The top 8 engines qualify for the next stage!

Confirmation of participation:
1. Andscacs
2. Vajolet
3. Fruit
4. Laser
5. Texel
6. Stockfish
7. Jonny
8. Fire
9. Nirvana
10. Komodo
11. Chiron
12. Fizbo

Best posts of the month