chessgod101
Please login to view all of the forum content.

Stockfish Polyglot

Page 2 of 3 Previous  1, 2, 3  Next

Go down

Stockfish Polyglot - Page 2 Empty Asking for explanation between several variants

Post  kazimirovich on Wed Feb 13, 2019 4:26 pm

Hello, first all all thx for this thread.

I have no idea on what are the differences between geral varial, modern, etc.

If someone could explain would be great. Thx in advance.

kazimirovich

Posts : 11
Points : 6
Reputation : -1
Join date : 2016-06-29

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Sat Feb 23, 2019 5:04 am

I guess it's all related to the computer and the processor you have: for Haswells use BMI2, for Xeons use POPCNT, for new cpus use MODERN, for old computers use OLD, et cetera.

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Sat Feb 23, 2019 8:40 pm

Stockfish Polyglot 2019-02-23.

This distribution is the expansion of Stockfish beta having timestamp: 1549618288 - Add KNNvKP Endgame Heuristic.

Author: Kurtbusch
Date: Thu Feb 21 19:53:03 2019 +0100
Timestamp: 1550775183

Add KNNvKP Endgame Heuristic

This is a somewhat different patch. It fixes blindspots for two knights vs pawn endgame.

With local testing starting from random KNNvKP positions where the pawn has not advanced beyond the 4th rank (thanks @protonspring !) at 15+0.15 (4 cores), this went +105=868-27 against master. All except two losses were won in reverse.

The heuristic is simple but effective - the strategy in these endgames is to push the opposing king to the corner, then move the knight that's blocking the pawn in for the checkmate while the pawn is free to move and prevents stalemate. This patch gives SF the little boost it needs to search the relevant king-cornering mating lines.

See the discussion in pull request 1939 for some more good results for this test in independant tests:
https://github.com/official-stockfish/Stockfish/pull/1939%22

Bench: 3310239


Author: MJZ1977
Date: Thu Feb 21 19:36:48 2019 +0100
Timestamp: 1550774208

ProbCutCount limit dependancy to cutNode

Use the ProbCutCount limit `2 + 2 * cutNode` instead of constant 3.

STC
LLR: -2.95 (-2.94,2.94) [0.50,4.50]
Total: 61812 W: 13599 L: 13459 D: 34754 Elo +0.79
http://tests.stockfishchess.org/tests/view/5c6d19240ebc5925cffca07a%22

LTC LLR: 2.96 (-2.94,2.94) [0.00,3.50]
Total: 27549 W: 4614 L: 4363 D: 18572 Elo +3.17
http://tests.stockfishchess.org/tests/view/5c6d45c10ebc5925cffca7a6%22

Closes https://github.com/official-stockfish/Stockfish/pull/2015%22 .

Bench: 3368889

Author: protonspring
Date: Thu Feb 21 19:24:02 2019 +0100
Timestamp: 1550773442

Remove PvNode dimension from Reductions array

This is a functional simplification: if we simply subtract one to Reductions[]
when PvNode is set, we can remove this dimension of the multidimensional array.
I think this saves about 8K of memory.

STC
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 10118 W: 2282 L: 2138 D: 5698 Elo +4.95
http://tests.stockfishchess.org/tests/view/5c6332b60ebc5925cffbdfed%22

LTC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 70765 W: 11617 L: 11575 D: 47573 Elo +0.21
http://tests.stockfishchess.org/tests/view/5c63379e0ebc5925cffbe0de%22

Closes https://github.com/official-stockfish/Stockfish/pull/2010%22 .

Bench 3261078

Author: protonspring
Date: Thu Feb 21 19:18:02 2019 +0100
Timestamp: 1550773082

Remove skipQuiets variable in search()

This is a functional simplification. The moveCountPruning variable and the skipQuiets variable are similar enough in function that they can be combined. This removes the skipQuiets variable in search.

STC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 23278 W: 5210 L: 5091 D: 12977 Elo +1.78
http://tests.stockfishchess.org/tests/view/5c65dc490ebc5925cffc12e9%22

LTC
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 77107 W: 12792 L: 12761 D: 51554 Elo +0.14
http://tests.stockfishchess.org/tests/view/5c65e4360ebc5925cffc1490%22

Closes https://github.com/official-stockfish/Stockfish/pull/2011%22 .

bench 3640330


Binaries of the latest version:

Windows:

- Stockfish_Polyglot_2019-02-23_32bit.exe
- Stockfish_Polyglot_2019-02-23_32bit_general.exe
- Stockfish_Polyglot_2019-02-23_32bit_old.exe
- Stockfish_Polyglot_2019-02-23_32bit_popc.exe
- Stockfish_Polyglot_2019-02-23_x64.exe
- Stockfish_Polyglot_2019-02-23_x64_bmi2.exe
- Stockfish_Polyglot_2019-02-23_x64_general.exe
- Stockfish_Polyglot_2019-02-23_x64_modern.exe
- Stockfish_Polyglot_2019-02-23_x64_popc.exe


Linux:

- Stockfish_Polyglot_2019-02-23_32bit
- Stockfish_Polyglot_2019-02-23_32bit_general
- Stockfish_Polyglot_2019-02-23_32bit_old
- Stockfish_Polyglot_2019-02-23_32bit_popc
- Stockfish_Polyglot_2019-02-23_x64
- Stockfish_Polyglot_2019-02-23_x64_bmi2
- Stockfish_Polyglot_2019-02-23_x64_general
- Stockfish_Polyglot_2019-02-23_x64_modern
- Stockfish_Polyglot_2019-02-23_x64_popc

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kazimirovich on Mon Feb 25, 2019 11:04 pm

kramnik wrote:I guess it's all related to the computer and the processor you have: for Haswells use BMI2, for Xeons use POPCNT, for new cpus use MODERN, for old computers use OLD, et cetera.

Great, thx for the information. Guess I have to use the modern one, as I have an AMD Athlon X4 860K Quad Core Processor. Pls, correct me if I'm wrong.

kazimirovich

Posts : 11
Points : 6
Reputation : -1
Join date : 2016-06-29

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Mon Feb 25, 2019 11:07 pm

No, you are right.

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Thu Mar 07, 2019 1:38 pm

Stockfish Polyglot: Updated March, 7 2019

- Matching official Stockfish beta having timestamp : 1549618288 - Add KNNvKP Endgame Heuristic. - ONLY FOR WINDOWS! NO LINUX, NO ANDROID, NO MACOS.

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Fri Mar 29, 2019 7:39 am

Stockfish Polyglot: Updated March, 29 2019.

Changelog:

This distribution is the expansion of Stockfish beta having timestamp: 1553445685 - Simplify pawn asymmetry (remove use of semiopen files). (#2054)

This is a functional simplification.

To me, the exclusive OR of semiopenFiles here is quite convoluted. Looks like it can be removed.

STC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 43885 W: 9731 L: 9653 D: 24501 Elo +0.62
http://tests.stockfishchess.org/tests/view/5c9041680ebc5925cfff10ea

LTC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 68437 W: 11577 L: 11533 D: 45327 Elo +0.22
http://tests.stockfishchess.org/tests/view/5c9101740ebc5925cfff1cbf

bench 3575627

Author: Joost VandeVondele
Date: Sun Mar 24 17:40:29 2019 +0100
Timestamp: 1553445629

Remove unneeded condition. (#2057)

This is covered by the line just before. If we would like to protect
against the piece value of e.g. a N == B, this could be done by an
assert, no need to do this at runtime.

No functional change.

Author: protonspring
Date: Sun Mar 24 17:37:38 2019 +0100
Timestamp: 1553445458

Simplify Passed Pawns (#2058)

This is a non-functional simplification/speedup.

The truth-table for popcount(support) >= popcount(lever) - 1 is:
------------------lever
------------------0-------1---------2
support--0------X-------X---------0
-----------1------X-------X---------X
-----------2------X-------X---------X

Thus, it is functionally equivalent to just do: support || !more_than_one(lever) which removes the expensive popcounts and the -1.

Result of 20 runs:
base (...h_master.exe) = 1451680 +/- 8202
test (./stockfish ) = 1454781 +/- 8604
diff = +3101 +/- 931

STC
LLR: 2.94 (-2.94,2.94) [-3.00,1.00]
Total: 35424 W: 7768 L: 7674 D: 19982 Elo +0.92
Http://tests.stockfishchess.org/tests/view/5c970f170ebc5925cfff5e28

No functional change.

Author: xoto10
Date: Wed Mar 20 14:57:34 2019 +0100
Timestamp: 1553090254

Remove !extension check #2045

While looking at pruning using see_ge() (which is very valuable)
it became apparent that the !extension test is not adding any
value - simplify it away.

STC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 56843 W: 12621 L: 12569 D: 31653 Elo +0.32
http://tests.stockfishchess.org/tests/view/5c8588cb0ebc5925cffe77f4

LTC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 78622 W: 13223 L: 13195 D: 52204 Elo +0.12
http://tests.stockfishchess.org/tests/view/5c8611cc0ebc5925cffe7f86

Further work could be to optimize the remaining see_ge() test. The idea of less pruning at higher depths is valuable, but perhaps the test (-PawnValueEg * depth) can be improved.

Bench: 3188688

Author: CoffeeOne
Date: Wed Mar 20 14:50:41 2019 +0100
Timestamp: 1553089841

Skip skipping thread scheme (#1972)


Several simplification tests (all with the bounds [-3,1]) were run:
5+0.05 8 threads, failed very quickly:
http://tests.stockfishchess.org/tests/view/5c439a020ebc5902bb5d3970

20+0.2 8 threads, also failed, but needed a lot more games:
http://tests.stockfishchess.org/tests/view/5c44b1b70ebc5902bb5d4e34

60+0.6 8 threads passed:
http://tests.stockfishchess.org/tests/view/5c48bfe40ebc5902bca15325

60+0.6 4 threads passed:
http://tests.stockfishchess.org/tests/view/5c4b71a00ebc593af5d49904

No functional change.

Author: Marco Costalba
Date: Tue Mar 12 08:35:10 2019 +0100
Timestamp: 1552376110

Increase thread stack for OS X (#2035)

On OS X threads other than the main thread are created with a reduced stack
size of 512KB by default, this is dangerously low for deep searches, so
adjust it to TH_STACK_SIZE. The implementation calls pthread_create() with
proper stack size parameter.

Verified for no regression at STC enabling the patch on all platforms where
pthread is supported.

LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 50873 W: 9768 L: 9700 D: 31405 Elo +0.46

No functional change.

Author: protonspring
Date: Sun Mar 10 10:53:39 2019 +0100
Timestamp: 1552211619

Remove popcount16() (#2038)

This is a non-functional simplification / code-style change.

This popcount16 method does nothing but initialize the PopCnt16 arrays.

This can be done in a single bitset line, which is less lines and more clear. Performance for this code is moot.

No functional change.

Author: xoto10
Date: Sun Mar 10 10:47:42 2019 +0100
Timestamp: 1552211262

Simplify failedLow away #1986

FailedLow doesn't seem to add any value so remove it.

STC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 43915 W: 9682 L: 9604 D: 24629 Elo +0.62
http://tests.stockfishchess.org/tests/view/5c5339770ebc592fc7baef74

LTC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 58515 W: 9670 L: 9609 D: 39236 Elo +0.36
http://tests.stockfishchess.org/tests/view/5c53cc840ebc592fc7baf6c1

Ideas for further work:

Tune the values in the revised fallingEval calculation
Consider adding a term using delta, e.g. c * (delta - 20) as an indicator of eval instability

Bench: 3318033

Author: Marco Costalba
Date: Sat Mar 9 13:28:11 2019 +0100
Timestamp: 1552134491

Revert "Allowing singular extension in mate positions"

It was causing an assert: value > -VALUE_INFINITE
under some conditions.

See https://github.com/official-stockfish/Stockfish/issues/2036

Bench: 3318033

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Thu Apr 04, 2019 9:30 am

Stockfish Polyglot: Updated Avril, 4 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1554025656 - Assorted trivial cleanups 3/2019 (#2030).

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Sun Apr 14, 2019 3:33 am

Stockfish Polyglot: Updated Avril, 13 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1555069684 - Revert "Shuffle detection #2064".

The value max_threads = std::thread::hardware_concurrency(); has been added to ucioption.cpp so that the engine settings dialog box is set by default with the maximum number of threads.
This modification has been made by me being inspired by the Raubfisch default settings.

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Mon Apr 15, 2019 8:16 pm

Stockfish Polyglot: Updated Avril, 15 2019 and fixed.

This distribution is the expansion of Stockfish beta having timestamp: 1555069684 - Revert "Shuffle detection #2064".

- The bug on the UCI interface regarding the number of threads has been corrected (thanks to the user "engine ratings").
- The bug of the OwnBook checkbox not active has been corrected.


Changes since the last Stockfish Polyglot release:

Author: Joost VandeVondele
Date: Sat Apr 13 17:17:47 2019 +0200
Timestamp: 1555168667

Remove two useless assignments (#2093)

These variables are initialized before their use in the movepicker loop.

passed STC:
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 138732 W: 30727 L: 30838 D: 77167 Elo -0.28
http://tests.stockfishchess.org/tests/view/5cb07af40ebc5925cf012c32

No functional change.

Author: Joost VandeVondele
Date: Sat Apr 13 17:15:15 2019 +0200
Timestamp: 1555168515

Give penalty for all early quiets of prev. ply

passed STC:
LLR: 2.94 (-2.94,2.94) [-3.00,1.00]
Total: 32884 W: 7283 L: 7184 D: 18417 Elo +1.05
http://tests.stockfishchess.org/tests/view/5cacb1b20ebc5925cf00ce97

passed LTC:
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 22869 W: 3920 L: 3803 D: 15146 Elo +1.78
http://tests.stockfishchess.org/tests/view/5cacbd760ebc5925cf00cfce

Bench: 3723099

Author: protonspring
Date: Sat Apr 13 16:58:41 2019 +0200
Timestamp: 1555167521

Simplify Connected Pawn Scoring #2090

This is a functional simplification that simplifies
connected scoring of pawns.

STC
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 37472 W: 8318 L: 8228 D: 20926 Elo +0.83
http://tests.stockfishchess.org/tests/view/5cae74ef0ebc5925cf00f8a5

LTC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 43035 W: 7366 L: 7281 D: 28388 Elo +0.69
http://tests.stockfishchess.org/tests/view/5caea3b50ebc5925cf00fe1e

Bench: 3470173  

Author: Massimiliano Goi
Date: Sat Apr 13 23:30:04 2019 +0200

The value max_threads = std::thread::hardware_concurrency(); has been added to ucioption.cpp so that the
engine settings dialog box is set by default with the maximum number of threads.

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Sat Apr 27, 2019 2:22 am

Stockfish Polyglot: Updated Avril, 26 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1556128317 - Remove useless initializations (#2115)".

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  Ginelu on Sun Apr 28, 2019 5:11 pm

kramnik wrote:Stockfish Polyglot: Updated Avril, 26 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1556128317 - Remove useless initializations (#2115)".

Thank you very much! Even though Stockfish does not help me much in mailing, I continue to ask her for an analysis of about an hour. It's the best engine in the world and I can not ignore it completely. Even if he loses the matches with the engines using BOOK.bin, he manages to take many draws and even win.

Ginelu

Posts : 120
Points : 132
Reputation : 14
Join date : 2019-03-20

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Mon Apr 29, 2019 2:27 am

You are free to do as you wish, of course. But if I were you I would definitely use CorChess or Komodo 12.3 for correspondence chess... Besides if you have Fritz there is the Clouding Room where you can rent monstrous computer for analysis. I often rent the one with most cores (hawk) or ducu7: both of them have asmFish and they behave very well in analysis, while asmfish running on my computer have the same incoherence problems than Stockfish and Raubfisch.

There is not a specific rule in chess, the last word is up to you, I can only show you a path. Experiment and find your own best combination.

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  Ginelu on Mon Apr 29, 2019 5:21 am

kramnik wrote:You are free to do as you wish, of course. But if I were you I would definitely use CorChess or Komodo 12.3 for correspondence chess... Besides if you have Fritz there is the Clouding Room where you can rent monstrous computer for analysis. I often rent the one with most cores (hawk) or ducu7: both of them have asmFish and they behave very well in analysis, while asmfish running on my computer have the same incoherence problems than Stockfish and Raubfisch.

There is not a specific rule in chess, the last word is up to you, I can only show you a path. Experiment and find your own best combination.

You are perfectly right. In other words, each one stretches as his blanket is.
 I did a test of 89 games Stockfish 270419 - Stockfish Polyglot 260419 at 1m + 1s and I was surprised that Sockfish 270419 won 12-10, the remaining draws. I gave Stockfish Polyglot Book.bin Goi5.1. I expect to win Stockfish Polyglot ...

Ginelu

Posts : 120
Points : 132
Reputation : 14
Join date : 2019-03-20

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Mon Apr 29, 2019 5:46 am

There is a significant elo raising in Stockfish270419:

Total: 17133 W: 3952 L: 3701 D: 9480 Elo +5.09
http://tests.stockfishchess.org/tests/view/5cc3550b0ebc5925cf02dada

passed LTC
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 37316 W: 6470 L: 6188 D: 24658 Elo +2.63

So I'm not surprised so much... Anyway I'm curious on what book did you use for Stockfish270419...

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  Ginelu on Mon Apr 29, 2019 3:48 pm

kramnik wrote:There is a significant elo raising in Stockfish270419:

Total: 17133 W: 3952 L: 3701 D: 9480 Elo +5.09
http://tests.stockfishchess.org/tests/view/5cc3550b0ebc5925cf02dada

passed LTC
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 37316 W: 6470 L: 6188 D: 24658 Elo +2.63

So I'm not surprised so much... Anyway I'm curious on what book did you use for Stockfish270419...

Since Stockfish 270419 does not have the Book.bin option, I think it will refer to Book.ctg. This was Goi5.1, downloaded free of charge from your site and for which I want to thank you very much.

Ginelu

Posts : 120
Points : 132
Reputation : 14
Join date : 2019-03-20

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Tue Apr 30, 2019 11:30 am

So in other words you have just tested Goi 5.1 BIN vs Goi 5.1 CTG update 2... My books are not for engines testing as they are very long, unless you fix the book moves up to move #3 or something around that. If you plan to test engines you should use a very low number of book moves or a book very, very short.

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  Ginelu on Tue Apr 30, 2019 5:32 pm

kramnik wrote:So in other words you have just tested Goi 5.1 BIN vs Goi 5.1 CTG update 2... My books are not for engines testing as they are very long, unless you fix the book moves up to move #3 or something around that. If you plan to test engines you should use a very low number of book moves or a book very, very short.

I wanted to test at the same time Goi 5.1.ctg. At first I was somewhat disappointed, because I did not find variants longer than 14 moves. I insisted after you repeated that the variants are long. I found in the French Defense (V. Tarrasch) a variant of 77 mutants that ended with the mat. I was surprised to find that the variation went on in a totally unbalanced position: Queen, Rook and pawn vs Rook! In such an end, no computer is needed. The sequel is useful in a difficult ending, such as a Queen ending or a finale with the difficulty of a study. We found a variant until moving 59 and Immortal v17. Now it depends on who played those variations: a GM, a long-time match engine ...

Ginelu

Posts : 120
Points : 132
Reputation : 14
Join date : 2019-03-20

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Thu May 02, 2019 7:44 am

Stockfish Polyglot: Updated May, 2 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1556390826 - Allow for address sanitizer. (#2119).

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  Ginelu on Thu May 02, 2019 3:57 pm

kramnik wrote:Stockfish Polyglot: Updated May, 2 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1556390826 - Allow for address sanitizer. (#2119).

Thank you very much!

Ginelu

Posts : 120
Points : 132
Reputation : 14
Join date : 2019-03-20

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Wed May 15, 2019 5:16 pm

Stockfish Polyglot: Updated May, 15 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1557912187 - Update failedHighCnt rule #2063..

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Tue Jun 04, 2019 2:49 pm

Stockfish Polyglot: Updated June, 4 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1559306139 - Scale lazy threshold according to material. (#2170).

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Fri Jun 21, 2019 4:20 pm

Stockfish Polyglot: Updated June, 21 2019.

Changelog: This distribution is the expansion of Stockfish beta having timestamp 1561042322 - More bonus for free passed pawn.

WARNING: very consistent update with significant enhancement in quality of play. I suggest to anyone publishing tests with Stockfish Polyglot to repeat them with this latest version because of the substantial difference in Elo.

Binaries:

Stockfish_Polyglot_2019-06-21_32bit.exe
Stockfish_Polyglot_2019-06-21_32bit_general.exe
Stockfish_Polyglot_2019-06-21_32bit_old.exe
Stockfish_Polyglot_2019-06-21_32bit_popc.exe
Stockfish_Polyglot_2019-06-21_x64.exe
Stockfish_Polyglot_2019-06-21_x64_bmi2.exe
Stockfish_Polyglot_2019-06-21_x64_general.exe
Stockfish_Polyglot_2019-06-21_x64_modern.exe
Stockfish_Polyglot_2019-06-21_x64_popc.exe

Changes since the last Stockfish Polyglot release:

Author: Vizvezdenec
Date: Thu Jun 20 16:52:02 2019 +0200
Timestamp: 1561042322

More bonus for free passed pawn

Give even more bonus to passed pawn if adjacent squares to its path
are not attacked.

passed STC
http://tests.stockfishchess.org/tests/view/5d08c9b10ebc5925cf0a6630
LLR: 2.95 (-2.94,2.94) [0.50,4.50]
Total: 175197 W: 39859 L: 38816 D: 96522 Elo +2.07

passed LTC
http://tests.stockfishchess.org/tests/view/5d0ab8240ebc5925cf0a8fe4
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 92928 W: 16124 L: 15682 D: 61122 Elo +1.65

Bench: 3398333

Author: Miguel Lahoz
Date: Thu Jun 20 16:39:22 2019 +0200
Timestamp: 1561041562

Change multi-cut pruning condition

Use comparison of eval with beta to predict potential cutNodes. This
allows multi-cut pruning to also prune possibly mislabeled Pv and NonPv
nodes.

STC:
LLR: 2.95 (-2.94,2.94) [0.50,4.50]
Total: 54305 W: 12302 L: 11867 D: 30136 Elo +2.78
http://tests.stockfishchess.org/tests/view/5d048ba50ebc5925cf0a15e8

LTC:
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 189512 W: 32620 L: 31904 D: 124988 Elo +1.31
http://tests.stockfishchess.org/tests/view/5d04bf740ebc5925cf0a17f0

Normally I would think such changes are risky, specially for PvNodes,
but after trying a few other versions, it seems this version is more
sound than I initially thought.

Aside from this, a small funtional change is made to return
singularBeta instead of beta to be more consistent with the fail-soft
logic used in other parts of search.

=============================

How to continue from there ?

We could try to audit other parts of the search where the "cutNode"
variable is used, and try to use dynamic info based on heuristic
eval rather than on this variable, to check if the idea behind this
patch could also be applied successfuly.

Bench: 3503788

Author: VoyagerOne
Date: Thu Jun 20 16:22:45 2019 +0200
Timestamp: 1561040565

QuietPick Speed-up

Non-functional speedup: no need to generate, score, or sort quiet moves
if SkipQuiet is true. Thanks to @mstembera for his suggestion.

STC:
LLR: 2.95 (-2.94,2.94) [0.00,4.00]
Total: 27910 W: 6406 L: 6129 D: 15375 Elo +3.45
http://tests.stockfishchess.org/tests/view/5d07e0920ebc5925cf0a58a8

Closes https://github.com/official-stockfish/Stockfish/pull/2194

No functional change

Author: Joost VandeVondele
Date: Thu Jun 20 16:15:57 2019 +0200
Timestamp: 1561040157

Fix progress issue with shuffling extensions

Fixes issues #2126 and #2189 where no progress in rootDepth is made for particular fens:

8/8/3P3k/8/1p6/8/1P6/1K3n2 b - - 0 1
8/1r1rp1k1/1b1pPp2/2pP1Pp1/1pP3Pp/pP5P/P5K1/8 w - - 79 46

the cause are the shuffle extensions. Upon closer analysis, it appears that in these cases a shuffle extension is made for every node searched, and progess can not be made. This patch implements a fix, namely to limit the number of extensions relative to the number of nodes searched. The ratio employed is 1/4, which fixes the issues seen so far, but it is a heuristic, and I expect that certain positions might require an even smaller fraction.

The patch was tested as a bug fix and passed:

STC:
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 56601 W: 12633 L: 12581 D: 31387 Elo +0.32
http://tests.stockfishchess.org/tests/view/5d02b37a0ebc5925cf09f6da

LTC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 52042 W: 8907 L: 8837 D: 34298 Elo +0.47
http://tests.stockfishchess.org/tests/view/5d0319420ebc5925cf09fe57

Furthermore, to confirm that the shuffle extension in this form indeed still brings Elo, one more test at VLTC was performed:
VLTC:
LLR: 2.96 (-2.94,2.94) [0.00,3.50]
Total: 142022 W: 20963 L: 20435 D: 100624 Elo +1.29
http://tests.stockfishchess.org/tests/view/5d03630d0ebc5925cf0a011a

Bench: 3961247

Author: syzygy1
Date: Thu Jun 20 16:09:40 2019 +0200
Timestamp: 1561039780

Partial revert of "Assorted trivial cleanups 5/2019".

Since root_probe() and root_probe_wdl() do not reset all tbRank values if they fail,
it is necessary to do this in rank_root_move(). This fixes issue #2196.
Alternatively, the loop could be moved into both root_probe() and root_probe_wdl().

No functional change

Author: VoyagerOne
Date: Fri Jun 14 19:59:17 2019 +0200
Timestamp: 1560535157

Simplify SEE Pruning (#2191)

Simplify SEE Pruning
Note this should also be a speedup...
If givesCheck is extended we know (except for DC) that it will have a positive SEE. So this new logic will be triggered before doing another expensive SEE function.

STC:
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 24429 W: 5484 L: 5368 D: 13577 Elo +1.65
http://tests.stockfishchess.org/tests/view/5cffbccd0ebc5925cf09a154

LTC:
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 28428 W: 4873 L: 4765 D: 18790 Elo +1.32
http://tests.stockfishchess.org/tests/view/5d0015f60ebc5925cf09acb1

Bench: 3897263

Author: protonspring
Date: Fri Jun 14 08:22:02 2019 +0200
Timestamp: 1560493322

Remove backmost_sq (#2190)

This is a non-functional simplification.

backmost_sq and frontmost_sq are redundant. It seems quite clear to always use frontmost_sq and use the correct color.

Non functional change.

Author: Stefan Geschwentner
Date: Fri Jun 14 07:36:42 2019 +0200
Timestamp: 1560490602

Increase pawns cache (#2187)

Increase size of the pawns table by the factor 8. This decreases the number of recalculations of pawn structure information significantly (at least at LTC).

I have done measurements for different depths and pawn cache sizes.
First are given the number of pawn entry calculations are done (in parentheses is the frequency that a call to probe triggers a pawn entry calculation). The delta% are the percentage of less done pawn entry calculations in comparison to master

VSTC: bench 1 1 12
STC: bench 8 1 16
LTC: bench 64 1 20
VLTC: bench 512 1 24

VSTC STC LTC VLTC
master 82218(6%) 548935(6%) 2415422(7%) 9548071(7%)
pawncache*2 79859(6%) 492943(5%) 2084794(6%) 8275206(6%)
pawncache*4 78551(6%) 458758(5%) 1827770(5%) 7112531(5%)
pawncache*8 77963(6%) 439421(4%) 1649169(5%) 6128652(4%)

delta%(p2-m) -2.9% -10.2% -13.7% -13.3%
delta%(p4-m) -4.5% -16.4% -24.3% -25.5%
delta%(p8-m) -5.2% -20.0% -31.7% -35.8%

STC: (non-regression test because at STC the effect is smaller than at LTC)
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 22767 W: 5160 L: 5040 D: 12567 Elo +1.83
http://tests.stockfishchess.org/tests/view/5d00f6040ebc5925cf09c3e2

LTC:
LLR: 2.94 (-2.94,2.94) [0.00,4.00]
Total: 26340 W: 4524 L: 4286 D: 17530 Elo +3.14
http://tests.stockfishchess.org/tests/view/5d00a3810ebc5925cf09ba16

No functional change.

Author: VoyagerOne
Date: Mon Jun 10 00:26:47 2019 +0200
Timestamp: 1560119207

No DC prune in QS (#2185)

Don't prune discover checks in qSearch

STC:
LLR: 2.96 (-2.94,2.94) [0.50,4.50]
Total: 23176 W: 5320 L: 5039 D: 12817 Elo +4.21
http://tests.stockfishchess.org/tests/view/5cfbc9350ebc5925cf094ab3

LTC:
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 128428 W: 22222 L: 21679 D: 84527 Elo +1.47
http://tests.stockfishchess.org/tests/view/5cfbf0b70ebc5925cf094ebc

Bench: 3883245

Author: Marco Costalba
Date: Sun Jun 9 14:57:08 2019 +0200
Timestamp: 1560085028

Assorted trivial cleanups 5/2019

No functional change.

bench: 4178282

Author: Stefan Geschwentner
Date: Sun Jun 9 14:34:51 2019 +0200
Timestamp: 1560083691

Remove depth condition for ttPv (#2166)

Currently PV nodes with a depth <= 4 were ignored for ttPv. Now remove this constraint and use all PV nodes.

STC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 52209 W: 11755 L: 11694 D: 28760 Elo +0.41
http://tests.stockfishchess.org/tests/view/5cebc2d30ebc5925cf07b93a

LTC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 20874 W: 3689 L: 3568 D: 13617 Elo +2.01
http://tests.stockfishchess.org/tests/view/5cec01fc0ebc5925cf07c62d

Author: protonspring
Date: Sun Jun 9 14:33:34 2019 +0200
Timestamp: 1560083614

Simplify passed pawns. (#2159)

This is a functional simplification.

If all of the stoppers are levers, a simple pawn push passes.

STC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 41768 W: 9360 L: 9278 D: 23130 Elo +0.68
http://tests.stockfishchess.org/tests/view/5ce82ed60ebc5925cf073a79

LTC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 40463 W: 6964 L: 6875 D: 26624 Elo +0.76
http://tests.stockfishchess.org/tests/view/5ce87d0b0ebc5925cf07472b

Author: protonspring
Date: Sun Jun 9 14:31:16 2019 +0200
Timestamp: 1560083476

Remove a few file_of's (simplify adjacent_files_bb) #2171

This is a non-functional simplification that removes two file_of(s).

STC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 22030 W: 5106 L: 4984 D: 11940 Elo +1.92
http://tests.stockfishchess.org/tests/view/5cf028de0ebc5925cf0839e7

Author: protonspring
Date: Sun Jun 9 14:28:42 2019 +0200
Timestamp: 1560083322

Simplify WeakUnopposedPawn #2181

This is a functional simplification.

Moves WeakUnopposedPawn to pawns.cpp and remove piece dependency.

STC
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 8699 W: 2000 L: 1853 D: 4846 Elo +5.87
http://tests.stockfishchess.org/tests/view/5cf7721b0ebc5925cf08ee79

LTC
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 46605 W: 7969 L: 7890 D: 30746 Elo +0.59
http://tests.stockfishchess.org/tests/view/5cf7d5f70ebc5925cf08fa96

Author: VoyagerOne
Date: Sun Jun 9 14:27:50 2019 +0200
Timestamp: 1560083270

SEE Pruning Tweak (#2183)

Don't SEE prune any check extensions

STC (yellow):
LLR: -2.96 (-2.94,2.94) [0.50,4.50]
Total: 129934 W: 29390 L: 28905 D: 71639 Elo +1.30
http://tests.stockfishchess.org/tests/view/5cf6b1a70ebc5925cf08dedb

LTC:
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 102115 W: 17692 L: 17224 D: 67199 Elo +1.59
http://tests.stockfishchess.org/tests/view/5cf830710ebc5925cf090331

Author: Michael Chaly
Date: Sun Jun 9 14:26:53 2019 +0200
Timestamp: 1560083213

Advanced pawn pushes tweak (#2175)

passed STC
http://tests.stockfishchess.org/tests/view/5cf586ee0ebc5925cf08c0ed
LLR: 2.96 (-2.94,2.94) [0.50,4.50]
Total: 29496 W: 6718 L: 6406 D: 16372 Elo +3.68

passed LTC
http://tests.stockfishchess.org/tests/view/5cf59b630ebc5925cf08c343
LLR: 2.95 (-2.94,2.94) [0.00,3.50]
Total: 40778 W: 7057 L: 6765 D: 26956 Elo +2.49

original idea from early 2018 by @jerrydonaldwatson
Code slightly rewritten to be shorter and more logical, no functinal changes
compared to passed patch.

Author: protonspring
Date: Sun Jun 9 14:24:06 2019 +0200
Timestamp: 1560083046

Simplify Outposts #2176

This is a functional simplification. This is NOT the exact version that was tested. Beyond the testing, an assignment was removed and a piece changes for consistency.

Instead of rewarding ANY square past an opponent pawn as an "outpost," only use squares that are protected by our pawn. I believe this is more consistent with what the chess world calls an "outpost."

STC
LLR: 2.95 (-2.94,2.94) [-3.00,1.00]
Total: 23540 W: 5387 L: 5269 D: 12884 Elo +1.74
http://tests.stockfishchess.org/tests/view/5cf51e6d0ebc5925cf08b823

LTC
LLR: 2.94 (-2.94,2.94) [-3.00,1.00]
Total: 53085 W: 9271 L: 9204 D: 34610 Elo +0.44
http://tests.stockfishchess.org/tests/view/5cf5279e0ebc5925cf08b992

bench 3424592

Author: 31m059
Date: Sun Jun 9 14:19:07 2019 +0200
Timestamp: 1560082747

Simplify k-value for passers. Bench: 3854907 (#2182)

Stockfish evaluates passed pawns in part based on a variable k, which shapes the passed pawn bonus based on the number of squares between the current square and promotion square that are attacked by enemy pieces, and the number defended by friendly ones. Prior to this commit, we gave a large bonus when all squares between the pawn and the promotion square were defended, and if they were not, a somewhat smaller bonus if at least the pawn's next square was. However, this distinction does not appear to provide any Elo at STC or LTC.

Where do we go from here? Many promising Elo-gaining patches were attempted in the past few months to refine passed pawn calculation, by altering the definitions of unsafe and defended squares. Stockfish uses these definitions to choose the value of k, so those tests interact with this PR. Therefore, it may be worthwhile to retest previously promising but not-quite-passing tests in the vicinity of this patch.

STC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 42344 W: 9455 L: 9374 D: 23515 Elo +0.66
http://tests.stockfishchess.org/tests/view/5cf83ede0ebc5925cf0904fb

LTC:
LLR: 2.96 (-2.94,2.94) [-3.00,1.00]
Total: 69548 W: 11855 L: 11813 D: 45880 Elo +0.21
http://tests.stockfishchess.org/tests/view/5cf8698f0ebc5925cf0908c8

Bench: 3854907

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Sat Jun 29, 2019 11:08 pm

Stockfish Polyglot: Updated June, 29 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1561621596 - Bonus for double attacks on unsupported pawns.

I WON'T MAKE NAMES THIS TIME; THERE IS NOT ANY PROBLEM IF I USE THE STOCKFISH NAME FOR THIS PROGRAM, IN FACTS THIS IS STOCKFISH!

I HAVE JUST ADDED TO THE CODE THE POSSIBILITY TO READ POLYGLOT BOOKS, THAT'S ALL. ALSO, MARCO COSTALBA HELPED ME A COUPLE OF TIMES WITH MY PROJECT (AND I THANK HIM), SO DON'T BOTHER THAT I SHOULDN'T USE THE NAME OF STOCKFISH, I AM PROTECTED BY BOTH THE COPYRIGHT AND THE GPL LAWS, AS THE SOURCE IS PUBLIC AND I DON'T GET ANY ECONOMICAL BENEFIT BY MAKING STOCKFISH POLYGLOT.

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  kramnik on Tue Jul 09, 2019 4:46 am

Stockfish Polyglot: Updated July, 9 2019.

This distribution is the expansion of Stockfish beta having timestamp: 1561982874 - Smoothly change playing strength with skill level. (#2142).

This is the fastest release ever of Stockfish Polyglot, reaching 1551 milliseconds in the benchmark test for the 64bit version.

THE CONTEMPT HAS BEEN SET TO OFF AND 0 AS DEFAULT, SEE THE ARTICLE: The Contempt-Coherence factor on Stockfish.

kramnik

Posts : 262
Points : 626
Reputation : 346
Join date : 2018-06-28

Back to top Go down

Stockfish Polyglot - Page 2 Empty Re: Stockfish Polyglot

Post  Sponsored content


Sponsored content


Back to top Go down

Page 2 of 3 Previous  1, 2, 3  Next

Back to top


 
Permissions in this forum:
You cannot reply to topics in this forum