Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

Martin Thoresen
Posts: 1833
Joined: Thu Jun 22, 2006 12:07 am

Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Post by Martin Thoresen »

Hiarcs 10
Daydreamer 1.75 32-bit
The Baron 2.23 x64
N2 0.4 32-bit
Arasan 11.7 32-bit
Philou 3.1.4.1
Cerebro 3.03d

32 games each, 224 games total.
Top
Martin Thoresen
Posts: 1833
Joined: Thu Jun 22, 2006 12:07 am

Re: Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Post by Martin Thoresen »

After 28 of 224 games:

Code: Select all

Gaviota_0_74_41_4CPU_Gauntlet_CCRL  2010

Gaviota 0.74.41 x64 4CPU   - Hiarcs 10                    1.0 - 3.0    +0/=2/-2    25.00%
Gaviota 0.74.41 x64 4CPU   - Daydreamer 1.75 32-bit       1.0 - 3.0    +1/=0/-3    25.00%
Gaviota 0.74.41 x64 4CPU   - The Baron 2.23 x64           1.0 - 3.0    +1/=0/-3    25.00%
Gaviota 0.74.41 x64 4CPU   - N2 0.4 32-bit                3.0 - 1.0    +3/=0/-1    75.00%
Gaviota 0.74.41 x64 4CPU   - Arasan 11.7 32-bit           2.0 - 2.0    +2/=0/-2    50.00%
Gaviota 0.74.41 x64 4CPU   - Philou 3.1.4.1               3.5 - 0.5    +3/=1/-0    87.50%
Gaviota 0.74.41 x64 4CPU   - Cerebro 3.03d                3.0 - 1.0    +2/=2/-0    75.00%
Download games: 1-28
Top
User avatar
michiguel
Posts: 6401
Joined: Thu Mar 09, 2006 8:30 pm
Location: Chicago, Illinois, USA

Re: Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Post by michiguel »

Martin Thoresen wrote:After 28 of 224 games:

Code: Select all

Gaviota_0_74_41_4CPU_Gauntlet_CCRL  2010

Gaviota 0.74.41 x64 4CPU   - Hiarcs 10                    1.0 - 3.0    +0/=2/-2    25.00%
Gaviota 0.74.41 x64 4CPU   - Daydreamer 1.75 32-bit       1.0 - 3.0    +1/=0/-3    25.00%
Gaviota 0.74.41 x64 4CPU   - The Baron 2.23 x64           1.0 - 3.0    +1/=0/-3    25.00%
Gaviota 0.74.41 x64 4CPU   - N2 0.4 32-bit                3.0 - 1.0    +3/=0/-1    75.00%
Gaviota 0.74.41 x64 4CPU   - Arasan 11.7 32-bit           2.0 - 2.0    +2/=0/-2    50.00%
Gaviota 0.74.41 x64 4CPU   - Philou 3.1.4.1               3.5 - 0.5    +3/=1/-0    87.50%
Gaviota 0.74.41 x64 4CPU   - Cerebro 3.03d                3.0 - 1.0    +2/=2/-0    75.00%
Download games: 1-28
This is very interesting an encouraging. Every single of these engines are superior to Gaviota single core vs single core. If this trends maintains, it means that the multithreaded version of Gaviota is working really well.

in this game

Code: Select all

[Event "Gaviota_0_74_41_4CPU_Gauntlet_CCRL"]
[Site "Norway"]
[Date "2010.05.12"]
[Round "2.5"]
[White "Arasan 11.7 32-bit"]
[Black "Gaviota 0.74.41 x64 4CPU"]
[Result "1-0"]
[ECO "B84"]
[EventDate "2010.05.11"]
[Annotator "0.53;-0.14"]
[PlyCount "107"]
[EventType "simul"]
[Source "Thoresen"]
[TimeControl "40/1020:0/0:0/0"]

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Be3 e6 7.Be2 Qc7 8.O-O b5 9.
Ncxb5 axb5 10.Nxb5 Qd8 11.e5 dxe5 12.Qxd8+ Kxd8 13.Bb6+ Ke7 14.Nc7 Ra4 15.b3 
Nbd7 16.bxa4 Nxb6 17.a5 Nbd5 18.Nb5 Ne4 19.Rfe1 Kf6 20.Bc4 Nd6 21.Nxd6 Bxd6 22.
Reb1 e4 23.Bxd5 exd5 24.Rb6 Ke5 25.a6 Bg4 26.f4+ Ke6 27.Kf1 Kd7 28.Rab1 Bc7 29.
R6b5 Kc6 30.R5b3 Bxf4 31.Rb6+ Kc5 32.R6b5+ Kd6 33.Rb6+ Ke7 34.h3 Bd7 35.a7 Ra8 
36.Ra6 Kd8 37.Ke2 Kc7 38.Ra5 Bc6 39.Rb3 Be5 40.c3 f5 41.h4 g6 42.h5 gxh5 43.Kf2
f4 44.Ke2 h4 45.Raa3 f3+ 46.gxf3 exf3+ 47.Kxf3 h3 48.Ke3 h2 49.Rb1 Bg3 50.Ra1 
Re8+ 51.Kd2 Re1 52.Rb1 h1=Q 53.Ra1 Qg2+ 54.Kd3 1-0
It look like the result should be the opposite, unless there was a loss on time or something like that? but that will be really weird at this time control.

Miguel
Top
Martin Thoresen
Posts: 1833
Joined: Thu Jun 22, 2006 12:07 am

Re: Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Post by Martin Thoresen »

michiguel wrote: This is very interesting an encouraging. Every single of these engines are superior to Gaviota single core vs single core. If this trends maintains, it means that the multithreaded version of Gaviota is working really well.

in this game

Code: Select all

[Event "Gaviota_0_74_41_4CPU_Gauntlet_CCRL"]
[Site "Norway"]
[Date "2010.05.12"]
[Round "2.5"]
[White "Arasan 11.7 32-bit"]
[Black "Gaviota 0.74.41 x64 4CPU"]
[Result "1-0"]
[ECO "B84"]
[EventDate "2010.05.11"]
[Annotator "0.53;-0.14"]
[PlyCount "107"]
[EventType "simul"]
[Source "Thoresen"]
[TimeControl "40/1020:0/0:0/0"]

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Be3 e6 7.Be2 Qc7 8.O-O b5 9.
Ncxb5 axb5 10.Nxb5 Qd8 11.e5 dxe5 12.Qxd8+ Kxd8 13.Bb6+ Ke7 14.Nc7 Ra4 15.b3 
Nbd7 16.bxa4 Nxb6 17.a5 Nbd5 18.Nb5 Ne4 19.Rfe1 Kf6 20.Bc4 Nd6 21.Nxd6 Bxd6 22.
Reb1 e4 23.Bxd5 exd5 24.Rb6 Ke5 25.a6 Bg4 26.f4+ Ke6 27.Kf1 Kd7 28.Rab1 Bc7 29.
R6b5 Kc6 30.R5b3 Bxf4 31.Rb6+ Kc5 32.R6b5+ Kd6 33.Rb6+ Ke7 34.h3 Bd7 35.a7 Ra8 
36.Ra6 Kd8 37.Ke2 Kc7 38.Ra5 Bc6 39.Rb3 Be5 40.c3 f5 41.h4 g6 42.h5 gxh5 43.Kf2
f4 44.Ke2 h4 45.Raa3 f3+ 46.gxf3 exf3+ 47.Kxf3 h3 48.Ke3 h2 49.Rb1 Bg3 50.Ra1 
Re8+ 51.Kd2 Re1 52.Rb1 h1=Q 53.Ra1 Qg2+ 54.Kd3 1-0
It look like the result should be the opposite, unless there was a loss on time or something like that? but that will be really weird at this time control.

Miguel
You are absolutely right. I haven't had the time to play through all the games, but this game is clearly finished with the wrong result by the GUI.

I will edit the result manually and re-upload the zip-file.

And as you say, time losses shouldn't happen with a repeating time control.

Good luck with the next games, I think this should be an interesting gauntlet to follow!
Top
User avatar
michiguel
Posts: 6401
Joined: Thu Mar 09, 2006 8:30 pm
Location: Chicago, Illinois, USA

Re: Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Post by michiguel »

Martin Thoresen wrote:
michiguel wrote: This is very interesting an encouraging. Every single of these engines are superior to Gaviota single core vs single core. If this trends maintains, it means that the multithreaded version of Gaviota is working really well.

in this game

Code: Select all

[Event "Gaviota_0_74_41_4CPU_Gauntlet_CCRL"]
[Site "Norway"]
[Date "2010.05.12"]
[Round "2.5"]
[White "Arasan 11.7 32-bit"]
[Black "Gaviota 0.74.41 x64 4CPU"]
[Result "1-0"]
[ECO "B84"]
[EventDate "2010.05.11"]
[Annotator "0.53;-0.14"]
[PlyCount "107"]
[EventType "simul"]
[Source "Thoresen"]
[TimeControl "40/1020:0/0:0/0"]

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Be3 e6 7.Be2 Qc7 8.O-O b5 9.
Ncxb5 axb5 10.Nxb5 Qd8 11.e5 dxe5 12.Qxd8+ Kxd8 13.Bb6+ Ke7 14.Nc7 Ra4 15.b3 
Nbd7 16.bxa4 Nxb6 17.a5 Nbd5 18.Nb5 Ne4 19.Rfe1 Kf6 20.Bc4 Nd6 21.Nxd6 Bxd6 22.
Reb1 e4 23.Bxd5 exd5 24.Rb6 Ke5 25.a6 Bg4 26.f4+ Ke6 27.Kf1 Kd7 28.Rab1 Bc7 29.
R6b5 Kc6 30.R5b3 Bxf4 31.Rb6+ Kc5 32.R6b5+ Kd6 33.Rb6+ Ke7 34.h3 Bd7 35.a7 Ra8 
36.Ra6 Kd8 37.Ke2 Kc7 38.Ra5 Bc6 39.Rb3 Be5 40.c3 f5 41.h4 g6 42.h5 gxh5 43.Kf2
f4 44.Ke2 h4 45.Raa3 f3+ 46.gxf3 exf3+ 47.Kxf3 h3 48.Ke3 h2 49.Rb1 Bg3 50.Ra1 
Re8+ 51.Kd2 Re1 52.Rb1 h1=Q 53.Ra1 Qg2+ 54.Kd3 1-0
It look like the result should be the opposite, unless there was a loss on time or something like that? but that will be really weird at this time control.

Miguel
You are absolutely right. I haven't had the time to play through all the games, but this game is clearly finished with the wrong result by the GUI.

I will edit the result manually and re-upload the zip-file.

And as you say, time losses shouldn't happen with a repeating time control.

Good luck with the next games, I think this should be an interesting gauntlet to follow!
It gives me quite a bit of material to work on. This is interesting:
Gaviota had HIARCS on the ropes but wasted the advantage in the diff. color bishop ending.

[Event "Gaviota_0_74_41_4CPU_Gauntlet_CCRL"]
[Site "Norway"]
[Date "2010.05.12"]
[Round "3.1"]
[White "Gaviota 0.74.41 x64 4CPU"]
[Black "Hiarcs 10"]
[Result "1/2-1/2"]
[ECO "B57"]
[EventDate "2010.05.11"]
[Annotator "0.10;0.48"]
[PlyCount "292"]
[EventType "simul"]
[Source "Thoresen"]
[TimeControl "40/1020:0/0:0/0"]

1.e4 c5 2.Nf3 Nc6 3.Nc3 d6 4.d4 cxd4 5.Nxd4 Nf6 6.Bc4 Qb6 7.Ndb5 a6 8.Be3 Qa5
9.Nd4 Ne5 10.Nb3 Qb4 11.Bd3 Nxd3+ 12.cxd3 Be6 13.O-O g6 14.Nd4 Bd7 15.a3 Qa5
16.b4 Qd8 17.h3 Bg7 18.Rc1 Rc8 19.Kh1 Nh5 20.Qd2 O-O 21.Rfe1 f5 22.exf5 gxf5
23.Nf3 f4 24.Qa2+ Kh8 25.Bd2 e6 26.Ne4 Rxc1 27.Rxc1 Qe7 28.Rc7 b6 29.Ra7 d5 30.
Nc3 Bf6 31.Qb1 Rg8 32.Qf1 Qd6 33.d4 a5 34.Ne5 Be8 35.Qb1 Rg7 36.Ra8 Qf8 37.bxa5
bxa5 38.Qd3 Bxe5 39.dxe5 Rg8 40.Ne2 Bg6 41.Rxf8 Bxd3 42.Rxg8+ Kxg8 43.Nd4 a4
44.Nxe6 Kf7 45.Nc7 d4 46.e6+ Kf6 47.Bb4 Ng7 48.e7 Kf7 49.Bc5 Bc4 50.Bxd4 Ne8
51.Nxe8 Kxe8 52.Bc5 Be6 53.Kh2 Bf5 54.g3 fxg3+ 55.Kxg3 Bd7 56.f3 Kf7 57.f4 Kf6
58.Kf3 Kf5 59.Bb4 Be8 60.Bd6 Bc6+ 61.Kf2 Ke4 62.Kg3 Kf5 63.Bb4 Bd7 64.Kf3 Be8
65.Ke3 Bc6 66.Bc5 h5 67.h4 Kf6 68.Kd4 Ke6 69.Bb4 Bd7 70.Bc5 Bc6 71.Ke3 Kf6 72.
Kd3 Kf5 73.Bd6 Bb5+ 74.Ke3 Ke6 75.Bc5 Be8 76.Bb4 Bc6 77.Kd4 Bd7 78.Kd3 Kf5 79.
Bd6 Bb5+ 80.Ke3 Ke6 81.Bc5 Be8 82.Kf3 Bc6+ 83.Kg3 Kf5 84.Bb4 Bd7 85.Bd6 Ke6 86.
Bc5 Kf5 87.Bb4 Bc6 88.Ba5 Kf6 89.Bb4 Kf5 90.Bd6 Ke6 91.Bc5 Kf5 92.Bb4 Bd7 93.
Bd6 Ke6 94.Bc5 Kf5 95.Bb6 Ke6 96.Bd8 Bc6 97.Kf2 Kf5 98.Ke3 Be8 99.Kd4 Kxf4 100.
Ba5 Bf7 101.Kc5 Ke5 102.Kc6 Ke6 103.Bb4 Be8+ 104.Kc7 Bb5 105.Kd8 Kf7 106.Bc5
Be8 107.Be3 Bc6 108.Bb6 Bb5 109.Bd4 Bc6 110.Be5 Bb5 111.Bc3 Bc6 112.Be5 Bb5
113.Bc3 Bc6 114.Bd4 Bb5 115.Be5 Be8 116.Bd4 Bc6 117.Be3 Be8 118.Bf4 Bc6 119.Bd6
Bb5 120.Be5 Be8 121.Bf4 Bc6 122.Bd6 Bb5 123.Bb4 Bc6 124.Bc5 Be8 125.Bb4 Bc6
126.Ba5 Be8 127.Bd2 Bb5 128.Be3 Bc6 129.Bf2 Bb5 130.Bb6 Bc6 131.Be3 Be8 132.Ba7
Bc6 133.Bf2 Bb5 134.Bb6 Bc6 135.Bc7 Bb5 136.Bg3 Bc6 137.Bf4 Be8 138.Bd2 Bb5
139.Ba5 Be8 140.Be1 Bb5 141.Bg3 Bc6 142.Bh2 Be8 143.Bb8 Bc6 144.Ba7 Be8 145.Bc5
Bb5 146.Bg1 Be8 1/2-1/2

Here, white should move f3 to capture f4 with the bishop, giving away the useless pawn in e7. That is the only chance.

In fact, black may have to play f3! to easily draw.
[d]4k3/4P2p/4b3/2B5/p4p2/P6P/5PPK/8 b - - 3 53 bm f3

And here white should play f3
[d]4k3/4P2p/8/2B2b2/p4p2/P6P/5PPK/8 w - - 4 54 bm f3
Top
Martin Thoresen
Posts: 1833
Joined: Thu Jun 22, 2006 12:07 am

Re: Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Post by Martin Thoresen »

I am glad I can help, Miguel! I wish for Gaviota to be as strong as possible.

Standings after 56 of 224 games:

Code: Select all

Gaviota_0_74_41_4CPU_Gauntlet_CCRL  2010

Gaviota 0.74.41 x64 4CPU   - Hiarcs 10                    2.0 - 6.0    +1/=2/-5    25.00%
Gaviota 0.74.41 x64 4CPU   - Daydreamer 1.75 32-bit       1.0 - 7.0    +1/=0/-7    12.50%
Gaviota 0.74.41 x64 4CPU   - The Baron 2.23 x64           4.5 - 3.5    +4/=1/-3    56.25%
Gaviota 0.74.41 x64 4CPU   - N2 0.4 32-bit                4.5 - 3.5    +4/=1/-3    56.25%
Gaviota 0.74.41 x64 4CPU   - Arasan 11.7 32-bit           4.5 - 3.5    +3/=3/-2    56.25%
Gaviota 0.74.41 x64 4CPU   - Philou 3.1.4.1               5.0 - 3.0    +4/=2/-2    62.50%
Gaviota 0.74.41 x64 4CPU   - Cerebro 3.03d                5.5 - 2.5    +4/=3/-1    68.75%
Download games: 29-56
Top
Martin Thoresen
Posts: 1833
Joined: Thu Jun 22, 2006 12:07 am

Re: Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Post by Martin Thoresen »

Standings after 84 of 224 games:

Code: Select all

Gaviota_0_74_41_4CPU_Gauntlet_CCRL  2010

Gaviota 0.74.41 x64 4CPU   - Hiarcs 10                    3.0 - 9.0    +2/=2/-8    25.00%
Gaviota 0.74.41 x64 4CPU   - Daydreamer 1.75 32-bit       2.0 - 10.0   +2/=0/-10   16.67%
Gaviota 0.74.41 x64 4CPU   - The Baron 2.23 x64           6.5 - 5.5    +5/=3/-4    54.17%
Gaviota 0.74.41 x64 4CPU   - N2 0.4 32-bit                6.0 - 6.0    +5/=2/-5    50.00%
Gaviota 0.74.41 x64 4CPU   - Arasan 11.7 32-bit           5.0 - 7.0    +3/=4/-5    41.67%
Gaviota 0.74.41 x64 4CPU   - Philou 3.1.4.1               9.0 - 3.0    +8/=2/-2    75.00%
Gaviota 0.74.41 x64 4CPU   - Cerebro 3.03d                8.0 - 4.0    +6/=4/-2    66.67%
Download games: 57-84
Top
Martin Thoresen
Posts: 1833
Joined: Thu Jun 22, 2006 12:07 am

Re: Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Post by Martin Thoresen »

Standings after 112 of 224 games:

Code: Select all

Gaviota_0_74_41_4CPU_Gauntlet_CCRL  2010

Gaviota 0.74.41 x64 4CPU   - Hiarcs 10                    4.0 - 12.0    +3/=2/-11    25.00%
Gaviota 0.74.41 x64 4CPU   - Daydreamer 1.75 32-bit       4.0 - 12.0    +4/=0/-12    25.00%
Gaviota 0.74.41 x64 4CPU   - The Baron 2.23 x64           8.5 - 7.5     +7/=3/-6     53.13%
Gaviota 0.74.41 x64 4CPU   - N2 0.4 32-bit                8.5 - 7.5     +6/=5/-5     53.13%
Gaviota 0.74.41 x64 4CPU   - Arasan 11.7 32-bit           6.5 - 9.5     +3/=7/-6     40.63%
Gaviota 0.74.41 x64 4CPU   - Philou 3.1.4.1               13.0 - 3.0    +12/=2/-2    81.25%
Gaviota 0.74.41 x64 4CPU   - Cerebro 3.03d                9.5 - 6.5     +7/=5/-4     59.38%
Download games: 85-112
Top
Martin Thoresen
Posts: 1833
Joined: Thu Jun 22, 2006 12:07 am

Re: Gaviota 0.74.41 x64 4CPU - Gauntlet (CCRL 40/40)

Post by Martin Thoresen »

After needing to manually adjust the result in a few games, all games have been re-uploaded with the corrections. Miguel has been informed of this.

The table after 112 games now looks like this:

Code: Select all

Gaviota_0_74_41_4CPU_Gauntlet_CCRL  2010

Gaviota 0.74.41 x64 4CPU   - Hiarcs 10                    4.0 - 12.0    +3/=2/-11    25.00%
Gaviota 0.74.41 x64 4CPU   - Daydreamer 1.75 32-bit       4.0 - 12.0    +4/=0/-12    25.00%
Gaviota 0.74.41 x64 4CPU   - The Baron 2.23 x64           8.5 - 7.5     +7/=3/-6     53.13%
Gaviota 0.74.41 x64 4CPU   - N2 0.4 32-bit                8.5 - 7.5     +6/=5/-5     53.13%
Gaviota 0.74.41 x64 4CPU   - Arasan 11.7 32-bit           7.5 - 8.5     +4/=7/-5     46.88%
Gaviota 0.74.41 x64 4CPU   - Philou 3.1.4.1               13.0 - 3.0    +12/=2/-2    81.25%
Gaviota 0.74.41 x64 4CPU   - Cerebro 3.03d                9.5 - 6.5     +7/=5/-4     59.38%
Download games:
1-28
29-56
57-84
85-112
Top

Return to “Computer Chess Club: Tournaments and Matches”