Kayra 1.0

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

mehmet123
Posts: 697
Joined: Sun Jan 26, 2020 10:38 pm
Location: Turkey
Full name: Mehmet Karaman

Re: Kayra 1.0

Post by mehmet123 »

ChickenLogic wrote: Wed Dec 22, 2021 11:01 pm
-10 elo at fishtest LTC will never turn to +20 elo at 6 cores.
My test say +4 elo not +20 elo.

Even if Kayra 1.0 is weaker than Stockfish Dev , don't forget this is the first version. Even the future versions of Kayra will be weaker than Stockfish , I have nothing to lose. Because I don't make any money from this engine anyway.
https://PrivateLadyEscorts.com - Live Local Dating - No Verify - Anonymous Casual Dating - Chat Local Singles
User avatar
AlexChess
Posts: 1563
Joined: Sat Feb 06, 2021 8:06 am
Full name: Alex Morales

Re: Kayra 1.0

Post by AlexChess »

I like Kayra and, due to error margins on my tests, even 1.0 probably isn't weaker than SF-DEV!

If you can, experiment also adding a profile/pattern to offer pawns and pieces for positional advantage, initiative (like Carlsen does :D )

Best regards, Alex
Chess engines and dedicated chess computers fan since 1981 :D macOS Sequoia 16GB-512GB, Windows 11 & Ubuntu ARM64.
ProteusSF Dev Forum
sarona
Posts: 127
Joined: Tue Oct 29, 2019 4:14 pm
Location: Canada
Full name: Ron Doughie

Re: Kayra 1.0

Post by sarona »

Alex, I have been looking at the 144 Kayra 1.0 v6 pgns you posted a link to on another forum and had a question concerning the games against Lc0 0.28.0.

Is this a GPU version of Lc0?
[pgn][Event "SuperBlitz3+3-BanksiaGui"]
[Site "Rome"]
[Date "2021.12.20"]
[Round "1.26"]
[White "Lc0 v0.28.0"]
[Black "Kayra 1.0 v6"]
[Result "0-1"]
[ECO "B90"]
[PlyCount "117"]
[EventDate "2021.??.??"]
[EventType "blitz"]
[FEN "6k1/6p1/3p2b1/2bP3p/5B1P/5Q2/4N1PK/3q4 w - - 0 45"]
[TimeControl "40/150+1:0"]

45. Ng3 {
-0.15/8 14845 14729} Qg1+ {-1.78/25 3699 8667227} 46. Kh3 {-0.17/1 10 4047} Be8
{-1.83/27 3708 8973108} 47. Nxh5 {-0.06/10 10622 5847} Qh1+ {-1.74/27 5430
13428618} 48. Kg3 {-0.04/8 6827 8816} Qe1+ {-2.45/29 6080 15702694} 49. Kh3 {
-0.04/9 9888 10414} Qh1+ {-2.58/31 3700 9486090} 50. Kg3 {-0.12/11 2401 8747}
Bg1 {-2.75/28 4266 12024680} 51. Kg4 {-0.21/12 11609 9026} Bd7+ {-3.05/25 4140
10130953} 52. Kg5 {-0.24/1 10 6722} Bb6 {-3.36/24 3987 10005809} 53. Nxg7 {
-0.59/13 13250 8443} Bd8+ {M+9/62 3805 9543965} 54. Kg6 {-1.81/12 6686 8325}
Qb1+ {M+8/74 3820 11812023} 55. Kh5 {-6.49/11 4691 7698} Kh7 {M+7/80 4004
13038514} 56. Qg3 {-8.80/8 5145 10526} Bf6 {M+6/95 4010 15746214} 57. Qg4 {
-28.24/8 10334 14988} Qe4 {M+5/159 4077 24448717} 58. Qg5 {-57.05/6 1982 1230}
Qe2+ {M+4/245 1037 6257093} 59. g4 {50 adjudication by engines' scores -10.54/
5 184} 0-1

[/pgn]

45. Ng3 looks to be losing immediately here.
User avatar
AlexChess
Posts: 1563
Joined: Sat Feb 06, 2021 8:06 am
Full name: Alex Morales

Re: Kayra 1.0

Post by AlexChess »

sarona wrote: Sat Dec 25, 2021 3:26 am Alex, I have been looking at the 144 Kayra 1.0 v6 pgns you posted a link to on another forum and had a question concerning the games against Lc0 0.28.0.

Is this a GPU version of Lc0?
[pgn][Event "SuperBlitz3+3-BanksiaGui"]
[Site "Rome"]
[Date "2021.12.20"]
[Round "1.26"]
[White "Lc0 v0.28.0"]
[Black "Kayra 1.0 v6"]
[Result "0-1"]
[ECO "B90"]
[PlyCount "117"]
[EventDate "2021.??.??"]
[EventType "blitz"]
[FEN "6k1/6p1/3p2b1/2bP3p/5B1P/5Q2/4N1PK/3q4 w - - 0 45"]
[TimeControl "40/150+1:0"]

45. Ng3 {
-0.15/8 14845 14729} Qg1+ {-1.78/25 3699 8667227} 46. Kh3 {-0.17/1 10 4047} Be8
{-1.83/27 3708 8973108} 47. Nxh5 {-0.06/10 10622 5847} Qh1+ {-1.74/27 5430
13428618} 48. Kg3 {-0.04/8 6827 8816} Qe1+ {-2.45/29 6080 15702694} 49. Kh3 {
-0.04/9 9888 10414} Qh1+ {-2.58/31 3700 9486090} 50. Kg3 {-0.12/11 2401 8747}
Bg1 {-2.75/28 4266 12024680} 51. Kg4 {-0.21/12 11609 9026} Bd7+ {-3.05/25 4140
10130953} 52. Kg5 {-0.24/1 10 6722} Bb6 {-3.36/24 3987 10005809} 53. Nxg7 {
-0.59/13 13250 8443} Bd8+ {M+9/62 3805 9543965} 54. Kg6 {-1.81/12 6686 8325}
Qb1+ {M+8/74 3820 11812023} 55. Kh5 {-6.49/11 4691 7698} Kh7 {M+7/80 4004
13038514} 56. Qg3 {-8.80/8 5145 10526} Bf6 {M+6/95 4010 15746214} 57. Qg4 {
-28.24/8 10334 14988} Qe4 {M+5/159 4077 24448717} 58. Qg5 {-57.05/6 1982 1230}
Qe2+ {M+4/245 1037 6257093} 59. g4 {50 adjudication by engines' scores -10.54/
5 184} 0-1

[/pgn]

45. Ng3 looks to be losing immediately here.
Is the CPU openblas build
Chess engines and dedicated chess computers fan since 1981 :D macOS Sequoia 16GB-512GB, Windows 11 & Ubuntu ARM64.
ProteusSF Dev Forum