Chess2u
Would you like to react to this message? Create an account in a few clicks or log in to continue.

Chess2uLog in

Kayra Tests:

descriptionKayra Tests: EmptyKayra Tests:

more_horiz
Kayra vs Stockfish:

Program                          Elo    +   -   Games   Score   Av.Op.  Draws

 1 Kayra 1.2 bmi2                 : 2402    5   4   640    50.5 %   2398   97.2 %
 2 Stockfish 190322 x64 bmi2      : 2398   4   5   640    49.5 %   2402   97.2  %


Individual statistics:

1 Kayra 1.2 bmi2            : 2402  640 (+ 12,=622,-  6), 50.5 %

Stockfish 190322 x64 bmi2     : 640 (+ 12,=622,-  6), 50.5 %

2 Stockfish 190322 x64 bmi2 : 2398  640 (+  6,=622,- 12), 49.5 %

Kayra 1.2 bmi2                : 640 (+  6,=622,- 12), 49.5 %


Game Conditions: Cutechess Gui, 1 Core (Core-i7 9750h), 5 min  TC, SuperGM 4 moves Opening Book, 128 Mb Hash, Ponder Off
mediafire.com file/4kzwg0wnb9z0nat/c7.pgn/file

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Kayra 1.2 vs Kayra 1.1:


Program Elo + - Games Score Av.Op. Draws

1 Kayra 1.2 bmi2 : 2406 8 7 640 51.7 % 2394 92.5 %
2 Kayra 1.1 bmi2 : 2394 7 8 640 48.3 % 2406 92.5 %


Individual statistics:

1 Kayra 1.2 bmi2 : 2406 640 (+ 35,=592,- 13), 51.7 %

Kayra 1.1 bmi2 : 640 (+ 35,=592,- 13), 51.7 %

2 Kayra 1.1 bmi2 : 2394 640 (+ 13,=592,- 35), 48.3 %

Kayra 1.2 bmi2 : 640 (+ 13,=592,- 35), 48.3 %


Game Conditions: Cutechess Gui, 1 Core (Core-i7 9750h), 5 min TC, SuperGM 4 moves Opening Book, 128 Mb Hash, Ponder Off
mediafire.com file/3ib3wgcx1ebx7ai/c8.pgn/file

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
I congratulate you for this strong engine but I do not know how you tested it
Kayra 1.2bmi2 VS
Stokfish 190322 BMI2
to me both engines are equal and this new engine Kayra 1.2a and the same but do not misunderstand me but on my pc are the same you hello spend a good day.

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Master 1 wrote:
I congratulate you for this strong engine but I do not know how you tested it
Kayra 1.2bmi2 VS
Stokfish 190322 BMI2
to me both engines are equal and this new engine Kayra 1.2a and the same but do not misunderstand me but on my pc are the same you hello spend a good day.


you right even for me sf more strong

and when i test many engines on my pc i fiend engine or 2 strong like sf but when i play vs some one using a non-engine sf i feel i well lose only sf play all game aggressive the other engines weakens over time

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Master 1 wrote:
I congratulate you for this strong engine but I do not know how you tested it
Kayra 1.2bmi2 VS
Stokfish 190322 BMI2
to me both engines are equal and this new engine Kayra 1.2a and the same but do not misunderstand me but on my pc are the same you hello spend a good day.



It is normal to have different results in engine tests if the chess engines are equal or close power.
Number of games, time control, opening books, ponder on/off etc. effect the results.

I mostly tested Kayra minimum 5 minutes/ game time control. Kayra isn' t so strong as Stockfish at bullet time control games in my tests.

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Kayra vs ProteusChessSF:

Program Elo + - Games Score Av.Op. Draws

1 Kayra 1.3 bmi2 : 2401 5 5 400 50.1 % 2400 97.8 %
2 ProteusChessSFv005 bmi2 : 2400 5 5 400 49.9 % 2401 97.8 %

Individual statistics:

1 Kayra 1.3 bmi2 : 2401 400 (+ 5,=391,- 4), 50.1 %

ProteusChessSFv005 bmi2 : 400 (+ 5,=391,- 4), 50.1 %

2 ProteusChessSFv005 bmi2 : 2400 400 (+ 4,=391,- 5), 49.9 %

Kayra 1.3 bmi2 : 400 (+ 4,=391,- 5), 49.9 %


Game Conditions: Cutechess Gui, 1 Core , Core-i7 9750h ( 6 cores/ 12 threads) , Concurrency:6, 5 min + 2 sec TC, SuperGM 4 Moves Opening Book, 128 Mb Hash, Ponder Off,
mediafire.com file/es5cfekd795o09r/c11.pgn/file

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Kayra 1.3 bmi2 - SF PB 190422 bmi2 / Blitz 5m+2s
Kayra Tests: Kayra-1-3-SF-PB-190422
https://mega.nz/file/6swFCCDa#PX8Z3QuqZ1T0JOqykNW6HQL0WtIeAT5JL4jVkIJC0-w
Openings: TCEC 21_Noomen Select
Processor: Intel ® Core ™ i7– 4790 CPU 3.60 GHz (4 Cores, 8 Threads)
Memoriy (RM): 32 GB, System type: 64-bit, Windows 10 PRO.
Graphic board: NVIDIA Geforce GTX 760

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
..............

Last edited by ChessFan1 on Mon Apr 25, 2022 1:37 am; edited 1 time in total

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
mehmet1921 wrote:
Kayra vs ProteusChessSF:

Program                          Elo    +   -   Games   Score   Av.Op.  Draws

 1 Kayra 1.3 bmi2                 : 2401    5   5   400    50.1 %   2400   97.8 %
 2 ProteusChessSFv005 bmi2        : 2400    5   5   400    49.9 %   2401   97.8 %

Individual statistics:

1 Kayra 1.3 bmi2            : 2401  400 (+  5,=391,-  4), 50.1 %

ProteusChessSFv005 bmi2       : 400 (+  5,=391,-  4), 50.1 %

2 ProteusChessSFv005 bmi2   : 2400 400 (+  4,=391,-  5), 49.9 %

Kayra 1.3 bmi2                : 400 (+  4,=391,-  5), 49.9 %


Game Conditions: Cutechess Gui, 1 Core , Core-i7 9750h ( 6 cores/ 12 threads) , Concurrency:6, 5 min + 2 sec TC, SuperGM 4 Moves Opening Book, 128 Mb Hash, Ponder Off,
mediafire.com file/es5cfekd795o09r/c11.pgn/file


Dear Mehmet, your engine is really strong! I prefer it to Stockfish and fear next Kayra 1.4 smile

https://www.chess2u.com/t18076p150-proteus-chess-sf-developing-my-stockfish-14-1-derivative#115323

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Kayra 1.3 bmi2 - CorChess 3 190422 bmi2 / Rapid 15m+10s
Kayra Tests: Kayra-1-3-Corchess190422
https://mega.nz/file/2pwk0AaL#mdHPTCBTC-Z2uE-Kcv6PkE737nO_lCHnVzow_-7uWzQ
Openings: TCEC 21_Noomen Select
Processor: Intel ® Core ™ i7– 4790 CPU 3.60 GHz (4 Cores, 8 Threads)
Memoriy (RM): 32 GB, System type: 64-bit, Windows 10.
Graphic board: NVIDIA Geforce GTX 760

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Sorry! The openings used in the match Kayra 1.3 bmi2 - CorChess 3 190422 bmi2 were Nunn 25 pgn, not TCEC.

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Kayra 1.3 bmi2 - SwordF PB 15.1_bmi2 / Blitz 5m+2s
Kayra Tests: Kayra-1-3-Sword-PB-15-1jpg
https://mega.nz/file/is43QCba#LaAnM02ZNz4QcAzcJ82jifVzS7SAZukG8zvwpCDn8js
Openings: TCEC 21_Noomen Select
Processor: Intel ® Core ™ i7– 4790 CPU 3.60 GHz (4 Cores, 8 Threads)
Memoriy (RM): 32 GB, System type: 64-bit, Windows 10.
Graphic board: NVIDIA Geforce GTX 760

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Kayra vs Stockfish/Stockfish Derivative Chess Engines (Ghppn Tests):

Program Elo + - Games Score Av.Op. Draws

1 Kayra 1.3 bmi2 : 2405 19 19 450 51.6 % 2395 64.9 %
2 CorChess 3 190422 : 2398 21 27 50 49.0 % 2405 94.0 %
3 SF PB 190422 : 2398 39 39 100 49.0 % 2405 68.0 %
4 ProteusSF JBE 007 bmi2 : 2398 45 45 100 49.0 % 2405 56.0 %
5 SwordF PB 15.1 : 2392 43 43 100 48.0 % 2405 60.0 %
6 ProteusChessSFv006 Win : 2388 43 43 100 47.5 % 2405 61.0 %


Individual statistics:

1 Kayra 1.3 bmi2 : 2405 450 (+ 86,=292,- 72), 51.6 %

ProteusChessSFv006 Win : 100 (+ 22,= 61,- 17), 52.5 %
SwordF PB 15.1 : 100 (+ 22,= 60,- 18), 52.0 %
CorChess 3 190422 : 50 (+ 2,= 47,- 1), 51.0 %
SF PB 190422 : 100 (+ 17,= 68,- 15), 51.0 %
ProteusSF JBE 007 bmi2 : 100 (+ 23,= 56,- 21), 51.0 %


Game Conditions: 5 min + 2 sec, Intel Core i7– 4790 (4 Cores, 8 Threads), TCEC 21Noomen Select
Game Conditions: (Kayra vs CorChess):15 min + 10 sec, Intel Core i7– 4790 (4 Cores, 8 Threads), TCEC 21 Noomen Select
https://pixeldrain.com/u/BBx6yeYq


Thanks Gheorghe for Kayra tests. Time condition in your tests ( 5 min + 2 sec) is so good. As ı wrote before " I think that Kayra will get much better results when tested with a sufficient number of games in rapid (more than 5 min/ game ) and long time games."

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
mehmet1921 wrote:
Kayra vs  Stockfish/Stockfish Derivative Chess Engines (Ghppn Tests):

Program                          Elo    +   -   Games   Score   Av.Op.  Draws

 1 Kayra 1.3 bmi2                 : 2405   19  19   450    51.6 %   2395   64.9 %
 2 CorChess 3 190422              : 2398   21  27    50    49.0 %   2405   94.0 %
 3 SF PB 190422                   : 2398  39  39   100    49.0 %   2405   68.0 %
 4 ProteusSF JBE 007 bmi2         : 2398   45  45   100    49.0 %   2405   56.0 %
 5 SwordF PB 15.1                 : 2392   43  43   100    48.0 %   2405   60.0 %
 6 ProteusChessSFv006 Win : 2388   43  43   100    47.5 %   2405   61.0 %


Individual statistics:

1 Kayra 1.3 bmi2            : 2405  450 (+ 86,=292,- 72), 51.6 %

ProteusChessSFv006 Win : 100 (+ 22,= 61,- 17), 52.5 %
SwordF PB 15.1                : 100 (+ 22,= 60,- 18), 52.0 %
CorChess 3 190422             :  50 (+  2,= 47,-  1), 51.0 %
SF PB 190422                  : 100 (+ 17,= 68,- 15), 51.0 %
ProteusSF JBE 007 bmi2        : 100 (+ 23,= 56,- 21), 51.0 %


Game Conditions: 5 min + 2 sec,  Intel  Core  i7– 4790 (4 Cores, 8 Threads), TCEC 21Noomen Select
Game Conditions: (Kayra vs CorChess):15 min + 10 sec,  Intel Core i7– 4790 (4 Cores, 8 Threads), TCEC 21 Noomen Select
https://pixeldrain.com/u/BBx6yeYq


Thanks Gheorghe for Kayra tests. Time  condition in your tests ( 5 min  + 2 sec)   is so good.  As  ı wrote before  " I think that Kayra will get much better results when tested with a sufficient number of games in rapid (more than 5 min/ game ) and long time games."


Thanks Gheorghe, I appreciate your statistical approach, I'm also trying do do the same

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
EMAN 7.93 64-bit BMI2 - Kayra 1.3 bmi2 / Blitz 5m+2s
Kayra Tests: Eman-7-93-kayra-1-3
https://mega.nz/file/flQHEQLZ#uopiLIpMUcAtEWrJGYEkhQ9T0bEF8pSlJ-1A6D4GA4w
Openings: TCEC 21_Noomen Select
Processor: Intel ® Core ™ i7– 4790 CPU 3.60 GHz (4 Cores, 8 Threads)
Memoriy (RM): 32 GB, System type: 64-bit, Windows 10 PRO.
Graphic board: NVIDIA Geforce GTX 760

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
im48 wrote:
EMAN 7.93 64-bit BMI2 - Kayra 1.3 bmi2 / Blitz 5m+2s
Kayra Tests: Eman-7-93-kayra-1-3
https://mega.nz/file/flQHEQLZ#uopiLIpMUcAtEWrJGYEkhQ9T0bEF8pSlJ-1A6D4GA4w
Openings: TCEC 21_Noomen Select
Processor: Intel ® Core ™ i7– 4790 CPU 3.60 GHz (4 Cores, 8 Threads)
Memoriy (RM): 32 GB, System type: 64-bit, Windows 10 PRO.
Graphic board: NVIDIA Geforce GTX 760


OK, I make an exception and test this not GPLv3 compliant also with Proteus on windows 11 ARM64, since I think that Kayra 1.3 although has not the latest patches is still ultra-strong, maybe as strong as Stockfish-dev 220422 with new net smile

Kind regards, Alex

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz


@Mehmet1921 you are right! ...It's already very difficult to have consistent data testing our engines  rolleyes  Deep Blue is an evident fake and the result absurd... lol

Last edited by AlexChess on Tue May 31, 2022 5:46 pm; edited 1 time in total

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz

Kayra vs Stockfish:

Program Elo + - Games Score Av.Op. Draws

1 Kayra 1.4 bmi2 : 2401 4 3 720 50.3 % 2399 98.1 %
2 Stockfish 210522 x64 bmi2 : 2399 3 4 720 49.7 % 2401 98.1 %


Individual statistics:

1 Kayra 1.4 bmi2 : 2401 720 (+ 9,=706,- 5), 50.3 %

Stockfish 210522 x64 bmi2 : 720 (+ 9,=706,- 5), 50.3 %

2 Stockfish 210522 x64 bmi2 : 2399 720 (+ 5,=706,- 9), 49.7 %

Kayra 1.4 bmi2 : 720 (+ 5,=706,- 9), 49.7 %


Game Conditions: Cutechess Gui, 1 Core , Core-i7 9750h ( 6 cores/ 12 threads) , Concurrency:6, 5 min + 1 sec TC, Balsa 5 Moves Opening Book, 128 Mb Hash, Ponder Off,


mediafire.com file/jdyr0vnjwqxix4h/c12.pgn/file

(mediafire.com/file/jdyr0vnjwqxix4h/c12.pgn/file)

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Welcome back Mehmet and Kayra 1.3 is still so strong... smile

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Kayra 1.4 bmi2 - SwordF PB 15.2 bmi2 / Blitz 5m+2s
Kayra Tests: Kayra-1-4-Sword-PB-15-2
https://mega.nz/file/r8Bw3ABQ#ExtXbPu9Lii4-sHQfLYzZBPy9A9wgWZaH8BgF7xfwsE
Openings: TCEC 22_Noomen 50 games
Processor: Intel ® Core ™ i7– 4790 CPU 3.60 GHz (4 Cores, 8 Threads)
Memoriy (RM): 32 GB, System type: 64-bit, Windows 10 PRO.
Graphic board: NVIDIA Geforce GTX 760

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Thx for a test. SwordF PB 15.2 and not Swordfish 15.2, does it mean you used a books in this test?

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
dorsz wrote:
Thx for a test. SwordF PB 15.2 and not Swordfish 15.2, does it mean you used a books in this test?


No, the settings of both engines were the default ones, without any additions. Under these conditions, most likely, SwordF PB 15.2 can have the same power as Swordfish 15.2, in no case weaker. This is why I use the PB version when I have it available.

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
im48 wrote:
dorsz wrote:
Thx for a test. SwordF PB 15.2 and not Swordfish 15.2, does it mean you used a books in this test?


No, the settings of both engines were the default ones, without any additions. Under these conditions, most likely, SwordF PB 15.2 can have the same power as Swordfish 15.2, in no case weaker. This is why I use the PB version when I have it available.


There may be cases where the compilations are weaker than the original, but here we also took into account the fact that Skynet is a very good compiler. His last compilation of Stockfish was the same as ChessMan's and that gives me confidence.
But don't worry, you made two top engines in a very short time and congratulations on that!

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
Dear Mehmet, I have compiled your wonderful Kayra 1.4 for Mac M1 (posting link on Talkchess in 30 mins http://talkchess.com/forum3/viewtopic.php?f=2&t=75911 and just testing it) I'm using Kayra 1.1.bin by Eduard Mehmet (I'm using Predator 2.2 always from him. Then I'll start tests with dorsz DOTS5 unbalanced test suites.

You are always welcome here: https://banksiagui.com/forums/viewtopic.php?p=132#p132 I'll soon publish all results and PGN.

descriptionKayra Tests: EmptyRe: Kayra Tests:

more_horiz
im48 wrote:
There may be cases where the compilations are weaker than the original, but here we also took into account the fact that Skynet is a very good compiler. His last compilation of Stockfish was the same as ChessMan's and that gives me confidence.


The differences are in speed only and they are really small differences. I'm using this tool for testing speed of my compilations.
https://github.com/zardav/FishBench

im48 wrote:
But don't worry, you made two top engines in a very short time and congratulations on that!

I did two because it was optimal number for me. Swordfish as more active playing and tactical engine than Stockfish being only slightly weaker, and Blue Marlin where the most important thing is its sacrificial style. Here is one of the latest results from a test tournament before I released Swordfish 15.2, a proof: (www.mediafire.com/view/b5hpqn5n6kt1w25/15.2vs15.1.png/file#)
I had stronger candidate of new Blue Marlin but I did not release it. The reason is it failed with tactical tests and it failed with the positions I verify the engine to have enough sacrificial style. Blue Marlin 15.1 was clearly better.
(My engines are top engines because they are Stockfish derivatives, none is stronger)

descriptionKayra Tests: EmptyRe: Kayra Tests:

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