in affine transform for AVX512/AVX2/SSSE3
The idea is to initialize sum with the first element instead of zero.
Reduce one add_epi32 and one set_zero SIMD instructions for each output dimension.
sum = 0; for i = 1 to n sum += a[i] ->
sum = a[1]; for i = 2 to n sum += a[i]
STC:
LLR: 2.95 (-2.94,2.94) {-0.25,1.25}
Total: 69048 W: 7024 L: 6799 D: 55225
Ptnml(0-2): 260, 5175, 23458, 5342, 289
https://tests.stockfishchess.org/tests/view/5faf2cf467cbf42301d6aa06
closes https://github.com/official-stockfish/Stockfish/pull/3227
No functional change.
This is a follow-up of the recent qsearch pruning patch in
a260c9a8a2
We now use the same guard condition (testing that we already have a defense with
a score better score than a TB loss) for all pruning heuristics in qsearch().
This allows some pruning when in check, but in a controlled way to ensure that
no wrong mate scores appear.
Tested with Elo-gaining bounds:
STC:
LLR: 2.97 (-2.94,2.94) {-0.25,1.25}
Total: 22632 W: 2433 L: 2264 D: 17935
Ptnml(0-2): 98, 1744, 7487, 1865, 122
https://tests.stockfishchess.org/tests/view/5fa59405936c54e11ec99515
LTC:
LLR: 2.94 (-2.94,2.94) {0.25,1.25}
Total: 105432 W: 4965 L: 4648 D: 95819
Ptnml(0-2): 85, 4110, 44011, 4423, 87
https://tests.stockfishchess.org/tests/view/5fa5b609936c54e11ec9952a
closes https://github.com/official-stockfish/Stockfish/pull/3221
Bench: 3578092
For the feature transformer the code is analogical to AVX2 since there was room for easy adaptation of wider simd registers.
For the smaller affine transforms that have 32 byte stride we keep 2 columns in one zmm register. We also unroll more aggressively so that in the end we have to do 16 parallel horizontal additions on ymm slices each consisting of 4 32-bit integers. The slices are embedded in 8 zmm registers.
These changes provide about 1.5% speedup for AVX-512 builds.
Closes https://github.com/official-stockfish/Stockfish/pull/3218
No functional change.
Using no searching time in case of a single legal move is not beneficial from
a strength point of view, and this special case can be easily removed:
STC:
LLR: 2.93 (-2.94,2.94) {-1.25,0.25}
Total: 22472 W: 2458 L: 2357 D: 17657
Ptnml(0-2): 106, 1733, 7453, 1842, 102
https://tests.stockfishchess.org/tests/view/5f926cbc81eda81bd78cb6df
LTC:
LLR: 2.94 (-2.94,2.94) {-0.75,0.25}
Total: 37880 W: 1736 L: 1682 D: 34462
Ptnml(0-2): 22, 1392, 16057, 1448, 21
https://tests.stockfishchess.org/tests/view/5f92a26081eda81bd78cb6fe
The advantage of using the normal time management for a single legal move is that scores
reported for that move are reasonable, not searching leads to artifacts during games
(see e.g. https://tcec-chess.com/#div=sf&game=96&season=19)
The disadvantage of using normal time management of a single legal move is that thinking
times can be unnaturally long, making it 'painful to watch' in online tournaments.
This patch uses normal time management, but caps the used time to 500ms.
This should lead to reasonable scores, and be hardly perceptible.
closes https://github.com/official-stockfish/Stockfish/pull/3195
closes https://github.com/official-stockfish/Stockfish/pull/3183
variant of a patch suggested by SFisGOD
No functional change.
Only do countermove based pruning in qsearch if we already have a move with a better score than a TB loss.
This patch fixes a bug (started as 843a961) that incorrectly prunes moves if in check,
and adds an assert to make sure no wrong mate scores are given in the future.
It replaces a no-op moveCount check with a check for bestValue.
Initially discussed in #3171 and later in #3199, #3198 and #3210.
This PR effectively closes#3171
It also likely fixes#3196 where this causes user visible incorrect TB scores,
which probably result from these incorrect mate scores.
Passed STC and LTC non-regression tests.
https://tests.stockfishchess.org/tests/view/5f9ef8dabca9bf35bae7f648
LLR: 2.93 (-2.94,2.94) {-1.25,0.25}
Total: 21672 W: 2339 L: 2230 D: 17103
Ptnml(0-2): 126, 1689, 7083, 1826, 112
https://tests.stockfishchess.org/tests/view/5f9f0caebca9bf35bae7f666
LLR: 2.97 (-2.94,2.94) {-0.75,0.25}
Total: 33152 W: 1551 L: 1485 D: 30116
Ptnml(0-2): 27, 1308, 13832, 1390, 19
closes https://github.com/official-stockfish/Stockfish/pull/3214
Bench: 3625915
A non-functional speedup. Unroll the loops going over
the output dimensions in the affine transform layers by
a factor of 4 and perform 4 horizontal additions at a time.
Instead of doing naive horizontal additions on each vector
separately use hadd and shuffling between vectors to reduce
the number of instructions by using all lanes for all stages
of the horizontal adds.
passed STC of the initial version:
LLR: 2.95 (-2.94,2.94) {-0.25,1.25}
Total: 17808 W: 1914 L: 1756 D: 14138
Ptnml(0-2): 76, 1330, 5948, 1460, 90
https://tests.stockfishchess.org/tests/view/5f9d516f6a2c112b60691da3
passed STC of the final version after cleanup:
LLR: 2.95 (-2.94,2.94) {-0.25,1.25}
Total: 16296 W: 1750 L: 1595 D: 12951
Ptnml(0-2): 72, 1192, 5479, 1319, 86
https://tests.stockfishchess.org/tests/view/5f9df5776a2c112b60691de3
closes https://github.com/official-stockfish/Stockfish/pull/3203
No functional change
This patch was inspired by c065abd which updates the accumulator,
if possible, based on the accumulator of two plies back if
the accumulator of the preceding ply is not available.
With this patch we look back even further in the position history
in an attempt to reduce the number of complete recomputations.
When we find a usable accumulator for the position N plies back,
we also update the accumulator of the position N-1 plies back
because that accumulator is most likely to be helpful later
when evaluating positions in sibling branches.
By not updating all intermediate accumulators immediately,
we avoid doing too much work that is not certain to be useful.
Overall, roughly 2-3% speedup.
This patch makes the code more specific to the net architecture,
changing input features of the net will require additional changes
to the incremental update code as discussed in the PR #3193 and #3191.
Passed STC:
https://tests.stockfishchess.org/tests/view/5f9056712c92c7fe3a8c60d0
LLR: 2.94 (-2.94,2.94) {-0.25,1.25}
Total: 10040 W: 1116 L: 968 D: 7956
Ptnml(0-2): 42, 722, 3365, 828, 63
closes https://github.com/official-stockfish/Stockfish/pull/3193
No functional change.
Idea of this patch can be described as following - in case we have consecutive fail highs and we reach late enough moves at root node probability of remaining quiet moves being able to produce even bigger value than moves that produced previous cutoff (so ones that should be high in move ordering but now they fail to produce beta cutoff because we actually reached high move count) should be quiet small so we can reduce them more.
passed STC
LLR: 2.94 (-2.94,2.94) {-0.25,1.25}
Total: 53392 W: 5681 L: 5474 D: 42237
Ptnml(0-2): 214, 4104, 17894, 4229, 255
https://tests.stockfishchess.org/tests/view/5f88501adcdad978fe8c527e
passed LTC
LLR: 2.94 (-2.94,2.94) {0.25,1.25}
Total: 59136 W: 2773 L: 2564 D: 53799
Ptnml(0-2): 30, 2117, 25078, 2300, 43
https://tests.stockfishchess.org/tests/view/5f884dbfdcdad978fe8c527a
closes https://github.com/official-stockfish/Stockfish/pull/3184
Bench: 4066972
We now include the total pawn count in the scaling factor for the output
of the NNUE evaluation network. This should have the effect of trying to
keep more pawns when SF has the advantage, but exchange them when she
is defending.
Thanks to Alexander Pagel (Lolligerhans) for the idea of using the
value of pawns to ease the comparison with the rest of the material
estimation.
Passed STC:
LLR: 2.93 (-2.94,2.94) {-0.25,1.25}
Total: 15072 W: 1700 L: 1539 D: 11833
Ptnml(0-2): 65, 1202, 4845, 1355, 69
https://tests.stockfishchess.org/tests/view/5f7235a63b22d6afa50699b3
Passed LTC:
LLR: 2.93 (-2.94,2.94) {0.25,1.25}
Total: 25880 W: 1270 L: 1124 D: 23486
Ptnml(0-2): 23, 980, 10788, 1126, 23
https://tests.stockfishchess.org/tests/view/5f723b483b22d6afa5069a99
closes https://github.com/official-stockfish/Stockfish/pull/3164
Bench: 3776081
Idea is that division by fraction of 2 is slightly faster than by other numbers so parameters are adjusted in a way that division in null move pruning depth reduction features dividing by 256 instead of dividing by 213.
Other than this patch is almost non-functional - difference starts to exist by depth 133.
passed STC
https://tests.stockfishchess.org/tests/view/5f70dd943b22d6afa50693c5
LLR: 2.95 (-2.94,2.94) {-0.25,1.25}
Total: 57048 W: 6616 L: 6392 D: 44040
Ptnml(0-2): 304, 4583, 18531, 4797, 309
passed LTC
https://tests.stockfishchess.org/tests/view/5f7180db3b22d6afa506941f
LLR: 2.95 (-2.94,2.94) {0.25,1.25}
Total: 45960 W: 2419 L: 2229 D: 41312
Ptnml(0-2): 43, 1779, 19137, 1987, 34
closes https://github.com/official-stockfish/Stockfish/pull/3159
bench 3789924
Current master uses a constant scale factor of 5/4 = 1.25 for the output
of the NNUE network, for compatibility with search and classical evaluation.
We modify this scale factor to make it dependent on the phase of the game,
going from about 1.5 in the opening to 1.0 for pure pawn endgames.
This helps Stockfish to avoid exchanges of pieces (heavy pieces in particular)
when she has the advantage, keeping more material on the board when attacking.
Passed STC:
LLR: 2.95 (-2.94,2.94) {-0.25,1.25}
Total: 14744 W: 1771 L: 1599 D: 11374
Ptnml(0-2): 87, 1184, 4664, 1344, 93
https://tests.stockfishchess.org/tests/view/5f6fb0a63b22d6afa506904f
Passed LTC:
LLR: 2.95 (-2.94,2.94) {0.25,1.25}
Total: 8912 W: 512 L: 393 D: 8007
Ptnml(0-2): 7, 344, 3637, 459, 9
https://tests.stockfishchess.org/tests/view/5f6fcf533b22d6afa5069066
closes https://github.com/official-stockfish/Stockfish/pull/3154
Bench: 3943952
- Clean signature of functions in namespace NNUE
- Add comment for countermove based pruning
- Remove bestMoveCount variable
- Add const qualifier to kpp_board_index array
- Fix spaces in get_best_thread()
- Fix indention in capture LMR code in search.cpp
- Rename TtmemDeleter to LargePageDeleter
Closes https://github.com/official-stockfish/Stockfish/pull/3063
No functional change
Use TT memory functions to allocate memory for the NNUE weights. This
should provide a small speed-up on systems where large pages are not
automatically used, including Windows and some Linux distributions.
Further, since we now have a wrapper for std::aligned_alloc(), we can
simplify the TT memory management a bit:
- We no longer need to store separate pointers to the hash table and
its underlying memory allocation.
- We also get to merge the Linux-specific and default implementations
of aligned_ttmem_alloc().
Finally, we'll enable the VirtualAlloc code path with large page
support also for Win32.
STC: https://tests.stockfishchess.org/tests/view/5f66595823a84a47b9036fba
LLR: 2.94 (-2.94,2.94) {-0.25,1.25}
Total: 14896 W: 1854 L: 1686 D: 11356
Ptnml(0-2): 65, 1224, 4742, 1312, 105
closes https://github.com/official-stockfish/Stockfish/pull/3081
No functional change.
