Stockfish

[Cr@ck]

Author: Shahin M. Shahin
Date: Sun May 26 20:32:41 2024 +0200
Timestamp: 1716748361

Remove rootDelta branch

This makes rootDelta logic easier to understand, recalculating the value
where it belongs so removes an unnecessary branch.

Passed non-regression STC:
https://tests.stockfishchess.org/tests/ ... d5e27d8d8e
LLR: 2.94 (-2.94,2.94) <-1.75,0.25>
Total: 206016 W: 53120 L: 53089 D: 99807 Elo +0.05
Ptnml(0-2): 591, 20928, 59888, 21061, 540

closes https://github.com/official-stockfish/S ... /pull/5289

No functional change

https://abrok.eu/stockfish/

From Stefan Pohl:

Wow ! This is not good...
”Latest Website-News (2024/06/02): Testrun of Stockfish 240526 finished: 0 Celo to Stockfish 240519, and sadly a measureable regression in EAS-scoring (nearly -20000 EAS points !)”

Thank you Stefan !

Can I have an explanatory team?

[dorsz]

Finaly LTC test of current version against the strongest one from May according to progression tests.
https://tests.stockfishchess.org/tests/ ... 1ad1c54f4b

[Homayoun]

Author: Tomasz Sobczyk
Date: Mon Jun 3 08:54:24 2024 +0200
Timestamp: 1717397664

Fix GetProcessGroupAffinity call

`GetProcessGroupAffinity` appears to require 4 byte alignment for `GroupArray` memory.

See https://stackoverflow.com/q/78567676 for further information

closes https://github.com/official-stockfish/S ... /pull/5340

No functional change


https://abrok.eu/stockfish/

[Homayoun]

Author: cj5716
Date: Sat Jun 8 23:40:22 2024 +0200
Timestamp: 1717882822

Make repeated bench runs identical

fixes https://github.com/official-stockfish/S ... ssues/5376
closes https://github.com/official-stockfish/S ... /pull/5377

No functional changes
https://abrok.eu/stockfish/

[Homayoun]

Author: FauziAkram
Date: Sat Jun 15 12:13:59 2024 +0200
Timestamp: 1718446439

Tweak the reduction formula

Tweak the reduction formula if position is or has been on the PV
Taking inspiration from an old Viren test.

Passed STC:
LLR: 2.94 (-2.94,2.94) <0.00,2.00>
Total: 78528 W: 20607 L: 20225 D: 37696 Elo +1.69
Ptnml(0-2): 262, 9297, 19785, 9637, 283
https://tests.stockfishchess.org/tests/ ... 4868d1fe24

Passed LTC:
LLR: 2.94 (-2.94,2.94) <0.50,2.50>
Total: 138630 W: 35666 L: 35132 D: 67832 Elo +1.34
Ptnml(0-2): 118, 15345, 37835, 15919, 98
https://tests.stockfishchess.org/tests/ ... 151f9e77b0

closes https://github.com/official-stockfish/S ... /pull/5385

Bench: 1134281


https://abrok.eu/stockfish/

[Homayoun]

Author: Tomasz Sobczyk
Date: Thu May 30 23:05:16 2024 +0200
Timestamp: 1717103116

Fix process' processor affinity determination on Windows.

Specialize and privatize NumaConfig::get_process_affinity.
Only enable NUMA capability for 64-bit Windows.

Following #5307 and some more testing it was determined that the way affinity
was being determined on Windows was incorrect, based on incorrect assumptions
about GetNumaProcessorNodeEx.

This patch fixes the issue by attempting to retrieve the actual process'
processor affinity using Windows API. However one issue persists that is not
addressable due to limitations of Windows, and will have to be considered a
limitation. If affinities were set using SetThreadAffinityMask instead of
SetThreadSelectedCpuSetMasks and GetProcessGroupAffinity returns more than 1
group it is NOT POSSIBLE to determine the affinity programmatically on Windows.
In such case the implementation assumes no affinites are set and will consider
all processors available for execution.

closes https://github.com/official-stockfish/S ... /pull/5312

No functional change

https://abrok.eu/stockfish/

[Homayoun]

Note:
With Excuse ,the last version is the previous one before my last sharing post.

[Homayoun]

Author: FauziAkram
Date: Sun Jun 23 11:46:41 2024 +0200
Timestamp: 1719136001

Internal iterative reductions: decrease depth more

For PV nodes without a ttMove, we decrease depth.
But in this patch, additionally, if the current position is found in the TT, and the stored depth in the TT is greater than or equal to
the current search depth, we decrease the search depth even further.

Passed STC:
LLR: 2.96 (-2.94,2.94) <0.00,2.00>
Total: 84384 W: 22154 L: 21761 D: 40469 Elo +1.62
Ptnml(0-2): 292, 9972, 21315, 10277, 336
https://tests.stockfishchess.org/tests/ ... 471b064db6

Passed LTC:
LLR: 2.95 (-2.94,2.94) <0.50,2.50>
Total: 92106 W: 23471 L: 23032 D: 45603 Elo +1.66
Ptnml(0-2): 79, 10155, 25154, 10578, 87
https://tests.stockfishchess.org/tests/ ... 471b064e56

closes https://github.com/official-stockfish/S ... /pull/5397

bench: 1038234

https://abrok.eu/stockfish/

[Sedat Canbaz]

Author: FauziAkram
Date: Sun Jun 23 11:46:41 2024 +0200
Timestamp: 1719136001

Internal iterative reductions: decrease depth more

For PV nodes without a ttMove, we decrease depth.
But in this patch, additionally, if the current position is found in the TT, and the stored depth in the TT is greater than or equal to
the current search depth, we decrease the search depth even further.

Passed STC:
LLR: 2.96 (-2.94,2.94) <0.00,2.00>
Total: 84384 W: 22154 L: 21761 D: 40469 Elo +1.62
Ptnml(0-2): 292, 9972, 21315, 10277, 336
https://tests.stockfishchess.org/tests/ ... 471b064db6

Passed LTC:
LLR: 2.95 (-2.94,2.94) <0.50,2.50>
Total: 92106 W: 23471 L: 23032 D: 45603 Elo +1.66
Ptnml(0-2): 79, 10155, 25154, 10578, 87
https://tests.stockfishchess.org/tests/ ... 471b064e56

closes https://github.com/official-stockfish/S ... /pull/5397

bench: 1038234

https://abrok.eu/stockfish/

As usual,
Many thanks to SF team and to Homayoun too

[Homayoun]

Thanks too, Sedat.