Alchemy

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

User avatar
Aser Huerga
Posts: 812
Joined: Tue Jun 16, 2009 10:09 am
Location: Spain

Alchemy

Post by Aser Huerga »

Alchemy

Hardware: Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHzDual 3000 MHz with 4.088 MB Memory
Operating system: Microsoft Windows XP 64 Bit Professional Service Pack 2 (Build 3790)
All engines 1 CPU
Arena 2.0.1 GUI
128 Hash Tables
3-4-5 EGTBs
Ponder off
Book GrandS PGN (by Norm Pollock) up to 8 moves, switching colours each pair
40 moves in 3 minutes repeating (adapted to the CCRL 40/4 standard)

36 rounds, 45 games each round, for a total of 1620 games.
Target engine: Deuterium 09.01.26.492

Participants

Code: Select all

Engine                     Score AmArBeDeDiEeGlGoKnRe    S-B
01: Amateur 2.82           0,0/0 ·                      0,00
01: Arion 1.7              0,0/0   ·                    0,00
01: Betsy 6.51 Nobook      0,0/0     ·                  0,00
01: Deuterium 09.01.26.492 0,0/0       ·                0,00
01: Dirty ACCA 3           0,0/0         ·              0,00
01: Eeyore 1.48            0,0/0           ·            0,00
01: Glass 1.3              0,0/0             ·          0,00
01: Gosu 0.16              0,0/0               ·        0,00
01: Knightx 1.92           0,0/0                 ·      0,00
01: Resp 0.19 x64_ja       0,0/0                   ·    0,00
I will post here some partial results and final standings. For a closer look: http://kirill-kryukov.com/chess/discuss ... 897#p49897
User avatar
Aser Huerga
Posts: 812
Joined: Tue Jun 16, 2009 10:09 am
Location: Spain

Re: Alchemy

Post by Aser Huerga »

Update: 30 rounds instead of 36, I won't be able to run some rounds owing to Christmas days, sorry.
Ferdy
Posts: 4851
Joined: Sun Aug 10, 2008 3:15 pm
Location: Philippines

Re: Alchemy

Post by Ferdy »

Aser Huerga wrote:Alchemy

Hardware: Intel(R) Core(TM)2 Duo CPU E8400 @ 3.00GHzDual 3000 MHz with 4.088 MB Memory
Operating system: Microsoft Windows XP 64 Bit Professional Service Pack 2 (Build 3790)
All engines 1 CPU
Arena 2.0.1 GUI
128 Hash Tables
3-4-5 EGTBs
Ponder off
Book GrandS PGN (by Norm Pollock) up to 8 moves, switching colours each pair
40 moves in 3 minutes repeating (adapted to the CCRL 40/4 standard)

36 rounds, 45 games each round, for a total of 1620 games.
Target engine: Deuterium 09.01.26.492

Participants

Code: Select all

Engine                     Score AmArBeDeDiEeGlGoKnRe    S-B
01: Amateur 2.82           0,0/0 ·                      0,00
01: Arion 1.7              0,0/0   ·                    0,00
01: Betsy 6.51 Nobook      0,0/0     ·                  0,00
01: Deuterium 09.01.26.492 0,0/0       ·                0,00
01: Dirty ACCA 3           0,0/0         ·              0,00
01: Eeyore 1.48            0,0/0           ·            0,00
01: Glass 1.3              0,0/0             ·          0,00
01: Gosu 0.16              0,0/0               ·        0,00
01: Knightx 1.92           0,0/0                 ·      0,00
01: Resp 0.19 x64_ja       0,0/0                   ·    0,00
I will post here some partial results and final standings. For a closer look: http://kirill-kryukov.com/chess/discuss ... 897#p49897
Hi Aser thanks for this test.

Ferdinand
User avatar
Aser Huerga
Posts: 812
Joined: Tue Jun 16, 2009 10:09 am
Location: Spain

Re: Alchemy

Post by Aser Huerga »

Hello Ferdinand,

I can say right now that you have done a fantastic job with this new version! By the moment, a ELO performance up to 2600 points!!

Standings after Round 4

Code: Select all

1   Deuterium 09.01.26.492 -2636-   27.5/36
2   Gosu 0.16               2486    23.0/36
3   Eeyore 1.48             2373    19.0/36
4   Betsy 6.51 Nobook       2459    18.5/36
5   Dirty ACCA 3            2470    18.0/36
6   Glass 1.3               2437    16.5/36  293.75
7   Knightx 1.92            2396    16.5/36  283.25
8   Amateur 2.82            2452    16.5/36  268.75
9   Arion 1.7               2427    12.5/36
10  Resp 0.19 x64_ja        2400    12.0/36
(All ELOs referred to the last Chess Computer Rating List when tournament starts, except -XXXX- one, wich is referred to performance).

30 rounds, 45 games each round, for a total of 1350 games.
Target engine: Deuterium 09.01.26.492

I will post here some partial results and final standings. For a closer look: http://kirill-kryukov.com/chess/discuss ... 897#p49897
User avatar
Aser Huerga
Posts: 812
Joined: Tue Jun 16, 2009 10:09 am
Location: Spain

Re: Alchemy

Post by Aser Huerga »

Standings after Round 8

Code: Select all

1   Deuterium 09.01.26.492 -2679-   58.0/72
2   Gosu 0.16               2486    41.5/72  1425.50
3   Glass 1.3               2437    41.5/72  1375.00
4   Dirty ACCA 3            2470    36.0/72
5   Eeyore 1.48             2373    34.5/72
6   Amateur 2.82            2452    34.0/72
7   KnightX 1.92            2396    32.5/72  1113.75
8   Betsy 6.51 Nobook       2459    32.5/72  1029.25
9   Arion 1.7               2427    29.5/72
10  Resp 0.19 x64_ja        2400    20.0/72
(All ELOs referred to the last Chess Computer Rating List when tournament starts, except -XXXX- one, wich is referred to performance).

30 rounds, 45 games each round, for a total of 1350 games.
Target engine: Deuterium 09.01.26.492

I will post here some partial results and final standings. For a closer look: http://kirill-kryukov.com/chess/discuss ... 897#p49897
User avatar
Aser Huerga
Posts: 812
Joined: Tue Jun 16, 2009 10:09 am
Location: Spain

Re: Alchemy

Post by Aser Huerga »

Standings after Round 12

Code: Select all

1   Deuterium 09.01.26.492 -2684-   87.5/108
2   Gosu 0.16               2486    61.5/108
3   Glass 1.3               2437    60.0/108
4   Dirty ACCA 3            2470    57.0/108
5   Amateur 2.82            2452    50.5/108
6   Betsy 6.51 Nobook       2459    50.0/108
7   Eeyore 1.48             2373    49.0/108
8   KnightX 1.92            2396    48.0/108
9   Arion 1.7               2427    45.5/108
10  Resp 0.19 x64_ja        2400    31.0/108
(All ELOs referred to the last Chess Computer Rating List when tournament starts, except -XXXX- one, wich is referred to performance).

30 rounds, 45 games each round, for a total of 1350 games.
Target engine: Deuterium 09.01.26.492

I will post here some partial results and final standings. For a closer look: http://kirill-kryukov.com/chess/discuss ... 897#p49897
Ferdy
Posts: 4851
Joined: Sun Aug 10, 2008 3:15 pm
Location: Philippines

Re: Alchemy

Post by Ferdy »

Aser Huerga wrote:Hello Ferdinand,

I can say right now that you have done a fantastic job with this new version! By the moment, a ELO performance up to 2600 points!!

Standings after Round 4

Code: Select all

1   Deuterium 09.01.26.492 -2636-   27.5/36
2   Gosu 0.16               2486    23.0/36
3   Eeyore 1.48             2373    19.0/36
4   Betsy 6.51 Nobook       2459    18.5/36
5   Dirty ACCA 3            2470    18.0/36
6   Glass 1.3               2437    16.5/36  293.75
7   Knightx 1.92            2396    16.5/36  283.25
8   Amateur 2.82            2452    16.5/36  268.75
9   Arion 1.7               2427    12.5/36
10  Resp 0.19 x64_ja        2400    12.0/36
(All ELOs referred to the last Chess Computer Rating List when tournament starts, except -XXXX- one, wich is referred to performance).

30 rounds, 45 games each round, for a total of 1350 games.
Target engine: Deuterium 09.01.26.492

I will post here some partial results and final standings. For a closer look: http://kirill-kryukov.com/chess/discuss ... 897#p49897
This version is seemed holding at this time control. I used to test this at 40 moves in 10 sec against 5 regular sparring partners, Yace, Hamsters, Hermann, Garbochess and Greko using cutechess-cli, and I got a rating of 2596, based on yace (ccrl 40/4) 2611 rating as reference.

Code: Select all

Rank Name                       Elo    +    - games score oppo. draws 
   1 Hamsters 0.7.1            2653   12   12  2180   59%  2583   22% 
   2 Yace 0.99.87              2611   13   13  1940   53%  2587   19% 
   3 Deuterium v09.02.27.001   2601   11   11  2800   58%  2543   21% 
   4 Deuterium v09.01.26.492   2596   11   11  2800   57%  2543   20% 
   5 Deuterium v09.02.27.002c  2588   12   12  2095   55%  2550   19% 
   6 Hermann 2.4               2567   12   12  2380   48%  2583   19% 
   7 Deuterium v09.02.27.002a  2563   12   12  2105   52%  2550   19% 
   8 Deuterium v09.02.27.002b  2541   25   25   491   49%  2545   19% 
   9 Deuterium v09.02.27.002   2448   60   64    84   29%  2602   20% 
  10 GarboChess 2.20           2442   13   14  1944   31%  2587   18% 
  11 GreKo 6.5                 2439   13   13  1931   30%  2587   20% 
User avatar
Aser Huerga
Posts: 812
Joined: Tue Jun 16, 2009 10:09 am
Location: Spain

Re: Alchemy

Post by Aser Huerga »

Hello Ferdinand,

Deuterium 09.01.26.492 ELO took me by surprise. I'm gonna start another tournament this time with opponents supposedly better than Deuterium, to specify its ELO. You know, the more opponents and the more games, the more accurate.

Best wishes.
User avatar
Graham Banks
Posts: 44662
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Alchemy

Post by Graham Banks »

Ferdy wrote:I used to test this at 40 moves in 10 sec against 5 regular sparring partners, Yace, Hamsters, Hermann, Garbochess and Greko using cutechess-cli, and I got a rating of 2596, based on yace (ccrl 40/4) 2611 rating as reference.
Hi Ferdinand,

based on the results I've seen, I'll get Deuterium to play a match against BBChess 1.3b or GarboChess 2.20 to see if it can make it into Division 3 of my next amateur series.

Cheers,
Graham.
gbanksnz at gmail.com
Ferdy
Posts: 4851
Joined: Sun Aug 10, 2008 3:15 pm
Location: Philippines

Re: Alchemy

Post by Ferdy »

Aser Huerga wrote:Hello Ferdinand,

Deuterium 09.01.26.492 ELO took me by surprise. I'm gonna start another tournament this time with opponents supposedly better than Deuterium, to specify its ELO. You know, the more opponents and the more games, the more accurate.

Best wishes.
I guess so, adding engines stronger than current Deuterium version would be appropriate.