PlentyChess 4.0.1

This is a bugfix release for the Tablebase functionality. Tablebases in version 4.0.0 mostly only work when analyzing FENs, and not during real games.

Binary Guide

• generic: Slowest
• ssse3: Slightly faster, requires a CPU with SSSE3 support
• fma: Faster, requires a CPU with FMA support
• avx2: Faster, requires a CPU with AVX2 support
• bmi2: Faster, requires a CPU with AVX2 and BMI2 support. Not recommended for Zen1 and Zen2 CPUs
• avx512: Faster, requires a CPU with AVX512 support
• avx512vnni: Fastest, requires a CPU with AVX512 and VNNI support
• neon: For ARM CPUs that support NEON instructions

If you're unsure about what binary to use, start with avx512vnni and find the first one that does not crash on your machine. There is a chance that the AVX2 binary is faster on your system, despite it supporting AVX512.

PlentyChess 4.0.0

This release includes the following notable improvements:

• Tuning for very high time controls
• Memory mapping of the neural network
• Various improvements to correction histories
• A different way of generating training data that makes the network play much better at high time controls
• Tablebase support

Performance vs. PlentyChess 3.0.0

STC

Elo   | 23.54 +- 2.99 (95%)
Conf  | 8.0+0.08s Threads=1 Hash=16MB
Games | N: 15006 W: 4298 L: 3283 D: 7425
Penta | [88, 1384, 3512, 2463, 56]
https://chess.aronpetkovski.com/test/7314/

LTC

Elo   | 41.52 +- 2.74 (95%)
Conf  | 40.0+0.40s Threads=1 Hash=64MB
Games | N: 15000 W: 4624 L: 2840 D: 7536
Penta | [25, 996, 3664, 2800, 15]
https://chess.aronpetkovski.com/test/7315/

Binary Guide

• generic: Slowest
• ssse3: Slightly faster, requires a CPU with SSSE3 support
• fma: Faster, requires a CPU with FMA support
• avx2: Faster, requires a CPU with AVX2 support
• bmi2: Faster, requires a CPU with AVX2 and BMI2 support. Not recommended for Zen1 and Zen2 CPUs
• avx512: Faster, requires a CPU with AVX512 support
• avx512vnni: Fastest, requires a CPU with AVX512 and VNNI support
• neon: For ARM CPUs that support NEON instructions

If you're unsure about what binary to use, start with avx512vnni and find the first one that does not crash on your machine. There is a chance that the AVX2 binary is faster on your system, despite it supporting AVX512.

PlentyChess 3.0.2

This release ships native ARM NEON binaries that offer up to 100% faster calculation speeds compared to version 3.0.1.
It is otherwise identical to version 3.0.1 and 3.0.0. Please download non-ARM binaries from release 3.0.1.

A big thanks to Darius from https://chessengeria.eu for helping and compiling the MacOS binary.

PlentyChess 3.0.1

This release contains bugfixes concerning the ucinewgame command for ChessBase and Fritz GUIs.
It is otherwise identical to version 3.0.0.

PlentyChess 3.0.0

This release is a huge milestone for PlentyChess. Previous versions used Lc0 data for training the networks, but the data used from this point forth is completely self-generated, with no connection to the previous networks.

Trained on 7.2 billion positions, the network in this release is estimated to be about 15 elo weaker than the Lc0 data network used up to version 2.1.0, but the much improved architecture and many many other improvements make this the strongest version of PlentyChess ever.

Performance vs. PlentyChess 2.0.0

All progression tests are run against PlentyChess 2.0.0, which is about 30 elo weaker than version 2.1.0.

STC

Elo   | 64.19 +- 2.64 (95%)
Conf  | 8.0+0.08s Threads=1 Hash=16MB
Games | N: 20020 W: 7075 L: 3418 D: 9527
Penta | [30, 1181, 4280, 4140, 379]
https://chess.aronpetkovski.com/test/6119/

LTC

Elo   | 55.93 +- 3.42 (95%)
Conf  | 40.0+0.40s Threads=1 Hash=64MB
Games | N: 10000 W: 3340 L: 1744 D: 4916
Penta | [4, 566, 2334, 2022, 74]
https://chess.aronpetkovski.com/test/6122/

DFRC

Elo   | 62.88 +- 6.66 (95%)
Conf  | 8.0+0.08s Threads=1 Hash=16MB
Games | N: 5016 W: 1827 L: 929 D: 2260
Penta | [62, 381, 925, 877, 263]
https://chess.aronpetkovski.com/test/6120/

Binary Guide

• generic: Slowest
• ssse3: Slightly faster, requires a CPU with SSSE3 support
• fma: Faster, requires a CPU with FMA support
• avx2: Faster, requires a CPU with AVX2 support
• bmi2: Faster, requires a CPU with AVX2 and BMI2 support. Not recommended for Zen1 and Zen2 CPUs
• avx512: Faster, requires a CPU with AVX512 support
• avx512vnni: Fastest, requires a CPU with AVX512 and VNNI support

If you're unsure about what binary to use, start with avx512vnni and find the first one that does not crash on your machine. There is a chance that the AVX2 binary is faster on your system, despite it supporting AVX512.

PlentyChess 2.1.0

This is mainly a bugfix release, since there is an issue with the UCI Chess960 / Ponder options in some GUIs.

Performance vs. PlentyChess 2.0.0

STC

Elo   | 27.49 +- 7.20 (95%)
Conf  | 8.0+0.08s Threads=1 Hash=16MB
Games | N: 2508 W: 685 L: 487 D: 1336
Penta | [12, 214, 605, 410, 13]
https://chess.aronpetkovski.com/test/1154/

LTC

Elo   | 31.33 +- 6.67 (95%)
Conf  | 40.0+0.40s Threads=1 Hash=64MB
Games | N: 2502 W: 697 L: 472 D: 1333
Penta | [5, 186, 640, 419, 1]
https://chess.aronpetkovski.com/test/1155/

Improvements

This version includes the following improvements compared to version 2.0.0:

• Correctly expose UCI "check" type options, fixing compatability with certain GUIs
• Stronger neural network, trained longer and on more data
• Much better thread voting, which makes use of the data of previous search iterations
• A big search and time management tune
• Improvements to move history updates

Binary guide

• generic: Slowest
• avx2: Faster, compatible with most modern CPUs
• bmi2: Faster, includes avx2, not recommended for Zen1/2 CPUs
• avx512: Fastest, includes bmi2

PlentyChess 2.0.0

Performance vs. PlentyChess 1.0.0

STC

Elo   | 61.20 +- 3.49 (95%)
Conf  | 8.0+0.08s Threads=1 Hash=16MB
Games | N: 20000 W: 6995 L: 3508 D: 9497
Penta | [40, 1356, 4318, 3649, 637]
https://chess.aronpetkovski.com/test/567/

LTC

Elo   | 49.14 +- 4.76 (95%)
Conf  | 40.0+0.40s Threads=1 Hash=64MB
Games | N: 10000 W: 3144 L: 1739 D: 5117
Penta | [2, 744, 2253, 1849, 152]
https://chess.aronpetkovski.com/test/568/

Improvements

This version includes the following improvements compared to version 1.0.0:

• Multithreading improvements due to thread voting
• Much stronger NNUE architecture featuring king buckets
• Many small improvements to the search algorithm and move selection
• UCI "Ponder" option
• Various bugfixes, including ones that were affecting strength

Binary guide

• generic: Slowest
• avx2: Faster, compatible with most modern CPUs
• bmi2: Faster, includes avx2, not recommended for Zen1/2 CPUs
• avx512: Fastest, includes bmi2

A big thank you to the members of my new OpenBench testing instance for our great collaboration.

PlentyChess 1.0.0

This is the biggest release of PlentyChess so far, shipping (D)FRC and MultiPV support.

Performance vs. PlentyChess 0.3.0

STC

Elo   | 139.43 +- 3.31 (95%)
Conf  | 8.0+0.08s Threads=1 Hash=16MB
Games | N: 25004 W: 11963 L: 2435 D: 10606
Penta | [20, 570, 3548, 6590, 1774]
https://openbench.yoshie2000.de/test/807/

LTC

Elo   | 137.16 +- 4.44 (95%)
Conf  | 40.0+0.40s Threads=1 Hash=64MB
Games | N: 12502 W: 5659 L: 965 D: 5878
Penta | [3, 196, 1697, 3814, 541]
https://openbench.yoshie2000.de/test/808/

Improvements

The following improvements have been made since version 0.3.0:

• Much stronger and larger neural network
• Various search improvements
• Much improved time management
• Many speedups
• More accurate eval normalization
• Added hashfull to the UCI output
• Added support for the UCI movestogo parameter
• Multiple fixes regarding zobrist keys, as well as other bugfixes

Binary guide

• generic: Slowest
• avx2: Faster, compatible with most modern CPUs
• bmi2: Faster, includes avx2, not recommended for Zen1/2 CPUs
• avx512: Fastest, includes bmi2

Note: The binaries include 2 bug fixes and have been built from commit 1d8b4fe

PlentyChess 0.3.0

This version features the following improvements:

• Lazy SMP
• Improvements to the transposition table
• Various search improvements & tunes
• Better neural network

Performance vs. PlentyChess 0.2.0

STC

Elo   | 153.01 +- 7.57 (95%)
Conf  | 8.0+0.08s Threads=1 Hash=16MB
Games | N: 5010 W: 2515 L: 441 D: 2054
Penta | [3, 100, 668, 1288, 446]
https://openbench.yoshie2000.de/test/419/

LTC

Elo   | 140.57 +- 10.13 (95%)
Conf  | 40.0+0.40s Threads=1 Hash=64MB
Games | N: 2506 W: 1161 L: 199 D: 1146
Penta | [0, 42, 348, 722, 141]
https://openbench.yoshie2000.de/test/420/

Lazy SMP Performance

4 threads (64mb Hash) vs. 1 Thread (16mb Hash)

Elo   | 119.89 +- 22.38 (95%)
Conf  | 8.0+0.08s Threads=4 Hash=64MB
Games | N: 500 W: 213 L: 47 D: 240
Penta | [0, 19, 65, 147, 19]
https://openbench.yoshie2000.de/test/421/

PlentyChess 0.2.1

Hotfix for the ucinewgame command. No further changes from 0.2.0.