Arasan 22.2

[jdart]

Arasan 22.2 is available from https://www.arasanchess.org

Changes in Arasan 22.2:
1) Cleanups and fixes in Makefile + CMakeLists.txt. Fix GCC version detection.
Better detection of clang compiler. Support Intel oneAPI complier.
2) Bug fixes in protocol class for cases where ponder search terminates early.
3) Make check extension always 1 ply + make it unconditional.
4) Bug fixes in hashing/retrieval of mate scores.
5) Changes to material imbalance scoring.
6) Fix bug preventing kibitzing of search info on chess servers.
7) Update to latest Fathom.
8) Bug fix: in MultiPV mode: apply correct aspiration window to search.
9) Support optimization for AVX2 capable processors.
10) Changes to time management code.
11) Stop scoring based on castling history, add instead "trapped rook" eval term
as in Stockfish.
12) Clean up/simplify parameter tuning code.
13) Fix improper hash table size computation for large hash sizes (>32G).
14) Changes to pruning code including null move.
15) Bug fix: Move check for forced move in search out of fail high/fail low loop.
16) Simplify/modify singular extension code. Add multi-cut pruning.
17) Add support for C++ 20 bit operations.

My testing indicates this version is only slightly ahead of 22.1 in terms of rating. Maybe some small improvement from AVX2 and other changes. But some of the bug fixes are significant.

--Jon

[Guenther]

Arasan 22.2 is available from https://www.arasanchess.org

Changes in Arasan 22.2:
1) Cleanups and fixes in Makefile + CMakeLists.txt. Fix GCC version detection.
Better detection of clang compiler. Support Intel oneAPI complier.
2) Bug fixes in protocol class for cases where ponder search terminates early.
3) Make check extension always 1 ply + make it unconditional.
4) Bug fixes in hashing/retrieval of mate scores.
5) Changes to material imbalance scoring.
6) Fix bug preventing kibitzing of search info on chess servers.
7) Update to latest Fathom.
8) Bug fix: in MultiPV mode: apply correct aspiration window to search.
9) Support optimization for AVX2 capable processors.
10) Changes to time management code.
11) Stop scoring based on castling history, add instead "trapped rook" eval term
as in Stockfish.
12) Clean up/simplify parameter tuning code.
13) Fix improper hash table size computation for large hash sizes (>32G).
14) Changes to pruning code including null move.
15) Bug fix: Move check for forced move in search out of fail high/fail low loop.
16) Simplify/modify singular extension code. Add multi-cut pruning.
17) Add support for C++ 20 bit operations.

My testing indicates this version is only slightly ahead of 22.1 in terms of rating. Maybe some small improvement from AVX2 and other changes. But some of the bug fixes are significant.

--Jon

Thanks for the christmas gift Jon!

[Graham Banks]

When I download the windows engine + book file, it has 0kb.

[jdart]

When I download the windows engine + book file, it has 0kb.

It's definitely on the server and is downloadable. You might want to check your antivirus. For example, Bitdefender may block the download, not because it detects malware, but because "this file is not commonly downloaded". But if I got to the download list in Firefox, it will let me allow the download.

--Jon

[Graham Banks]

When I download the windows engine + book file, it has 0kb.

It's definitely on the server and is downloadable. You might want to check your antivirus. For example, Bitdefender may block the download, not because it detects malware, but because "this file is not commonly downloaded". But if I got to the download list in Firefox, it will let me allow the download.

--Jon

Got it. Thanks.

[Jouni]

Arasan test suite is no more updated? I use it a lot. One buggy position is probably 121:

[d]8/5p1k/6p1/1p1Q3p/3P4/1R2P1KP/6P1/r4q2 b - - 0 1

Analysis by Stockfish 181220:

1. -+ (-11.83): 1...Qe1+ 2.Kf3 Qf1+ 3.Kg3 h4+ 4.Kxh4 Qf2+ 5.Kg4 Rf1 6.Ra3 Qe2+ 7.Kh4 Rf5 8.Qxf5 Qe1+ 9.Kg4 gxf5+ 10.Kf3 b4 11.Ra2 b3 12.Rb2 Qc1 13.Re2 Qf1+ 14.Rf2 Qd1+ 15.Kg3 Qe1 16.Kf3 f4 17.exf4 Qc3+ 18.Kg4 b2 19.Rf1 Qe3 20.Rh1 Qf2 21.g3 Qe2+ 22.Kh4 Qe7+ 23.Kg4 Qe4 24.Rg1 Qe2+ 25.Kf5 b1Q+ 26.Rxb1 Qe6+ 27.Kg5 Qg6+ 28.Kh4 Qxb1 29.g4 Qe4 30.Kg3 Qd3+ 31.Kg2 Qc2+ 32.Kg1 Qd2 33.f5
2. -+ (-11.45): 1...h4+ 2.Kxh4 Qf2+ 3.Kg4 Rf1 4.Ra3 Qe2+ 5.Kh4 Rf5 6.Qxf5 Qe1+ 7.Kg5 gxf5 8.Kf6 Qh4+ 9.Kxf7 Qd8 10.Ra1 b4 11.h4 b3 12.Rb1 Qc7+ 13.Kf6 Qc2 14.Rg1 b2 15.h5 Qc6+ 16.Ke5 Qe4+ 17.Kf6 Qxe3 18.Rd1 Qb3 19.Rh1 Qb7 20.h6 Qc6+ 21.Ke7 Qxg2 22.Re1 Qd2 23.Rf1

[Chessqueen]

Arasan 22.2 is available from https://www.arasanchess.org

Changes in Arasan 22.2:
1) Cleanups and fixes in Makefile + CMakeLists.txt. Fix GCC version detection.
Better detection of clang compiler. Support Intel oneAPI complier.
2) Bug fixes in protocol class for cases where ponder search terminates early.
3) Make check extension always 1 ply + make it unconditional.
4) Bug fixes in hashing/retrieval of mate scores.
5) Changes to material imbalance scoring.
6) Fix bug preventing kibitzing of search info on chess servers.
7) Update to latest Fathom.
8) Bug fix: in MultiPV mode: apply correct aspiration window to search.
9) Support optimization for AVX2 capable processors.
10) Changes to time management code.
11) Stop scoring based on castling history, add instead "trapped rook" eval term
as in Stockfish.
12) Clean up/simplify parameter tuning code.
13) Fix improper hash table size computation for large hash sizes (>32G).
14) Changes to pruning code including null move.
15) Bug fix: Move check for forced move in search out of fail high/fail low loop.
16) Simplify/modify singular extension code. Add multi-cut pruning.
17) Add support for C++ 20 bit operations.

My testing indicates this version is only slightly ahead of 22.1 in terms of rating. Maybe some small improvement from AVX2 and other changes. But some of the bug fixes are significant.

--Jon

NO NN or NNUE Support ?

[jdart]

>Arasan test suite is no more updated? I use it a lot. One buggy position is probably 121

I have collected some newer test positions but haven't put them into a new test suite, not yet. It is on my to-do list.

Meanwhile, thanks for the correction. I will do some testing on this one but it looks like you can add Qe1+ as an alternate solution.

--Jon

[jdart]

NO NN or NNUE Support ?

No, not yet. If you want it, you can submit a pull request .

--Jon

[Wilson]

Thanks for a new Arasan, Jon.

3) Do you mean that you now always extend checks even if see < 0? What's the Elo diff with see >= 0 only?