Super Tournament II (final standings after 52 rounds)

[Graham Banks]

Geschrieben von:/Posted by: Graham Banks at 27. June 2004 00:07:

Athlon 2000
128mb hash each
3,4,5 piece tablebases
All using Fritz Powerbook (maximum variety, minimum 5 games for a move to be played, no learning)
40 moves in 80 minutes repeating.
Ponder off
4 cycles (52 rounds)
Tournament run under Fritz GUI

FINAL STANDINGS (52 rounds)

Deep Fritz 8 - 35.0
Shredder 8 - 33.5
Hiarcs 9 - 31.5
Ruffian 2.1.0 - 27.5
CM9000 Judge - 27.5
Chess Tiger 15 - 27.5
Aristarch 4.50 - 26.5
Deep Junior 8 - 26.0
SmarThink 0.17a - 25.5
List 512 - 25.0
Deep Sjeng 1.6 - 22.0
El Chinito 3.25 - 21.5
Thinker 4.6b - 18.5
Crafty 19.13 - 16.5

[Niyaz Khasanov]

Geschrieben von:/Posted by: Niyaz Khasanov at 28 June 2004 17:09:22:
Als Antwort auf:/In reply to: Super Tournament II (final standings after 52 rounds) geschrieben von:/posted by: Graham Banks at 27. June 2004 00:07:

Athlon 2000
128mb hash each
3,4,5 piece tablebases
All using Fritz Powerbook (maximum variety, minimum 5 games for a move to be played, no learning)
40 moves in 80 minutes repeating.
Ponder off
4 cycles (52 rounds)
Tournament run under Fritz GUI
FINAL STANDINGS (52 rounds)

Deep Fritz 8 - 35.0
Shredder 8 - 33.5
Hiarcs 9 - 31.5
Ruffian 2.1.0 - 27.5
CM9000 Judge - 27.5
Chess Tiger 15 - 27.5
Aristarch 4.50 - 26.5
Deep Junior 8 - 26.0
SmarThink 0.17a - 25.5
List 512 - 25.0
Deep Sjeng 1.6 - 22.0
El Chinito 3.25 - 21.5
Thinker 4.6b - 18.5
Crafty 19.13 - 16.5

It is very interesting.
I have two questions.
1. Were all games of Aristarch made with the 4.50 version?
Or there is a composition of 4.50 and 4.41 games?
2. I have an idea that if one book is used with several engines,
the length of each line should be 5..10 moves.
It simulates the real tournaments, where engines with different books go out of their books early.
The noise because of book holes is minimized.
We can also measure the opening strength and the early middlegame strength
of engines.
What do you think about this idea?

[Graham Banks]

Geschrieben von:/Posted by: Graham Banks at 28. June 2004 21:59:
Als Antwort auf:/In reply to: Re: Super Tournament II (final standings after 52 rounds) geschrieben von:/posted by: Niyaz Khasanov at 28 June 2004 17:09:22:

Athlon 2000
128mb hash each
3,4,5 piece tablebases
All using Fritz Powerbook (maximum variety, minimum 5 games for a move to be played, no learning)
40 moves in 80 minutes repeating.
Ponder off
4 cycles (52 rounds)
Tournament run under Fritz GUI
FINAL STANDINGS (52 rounds)

Deep Fritz 8 - 35.0
Shredder 8 - 33.5
Hiarcs 9 - 31.5
Ruffian 2.1.0 - 27.5
CM9000 Judge - 27.5
Chess Tiger 15 - 27.5
Aristarch 4.50 - 26.5
Deep Junior 8 - 26.0
SmarThink 0.17a - 25.5
List 512 - 25.0
Deep Sjeng 1.6 - 22.0
El Chinito 3.25 - 21.5
Thinker 4.6b - 18.5
Crafty 19.13 - 16.5

It is very interesting.
I have two questions.
1. Were all games of Aristarch made with the 4.50 version?
Or there is a composition of 4.50 and 4.41 games?
2. I have an idea that if one book is used with several engines,
the length of each line should be 5..10 moves.
It simulates the real tournaments, where engines with different books go out of their books early.
The noise because of book holes is minimized.
We can also measure the opening strength and the early middlegame strength
of engines.
What do you think about this idea?

Aristarch 4.50 played all rounds.
Using the powerbook with the minimum set at 5 saw most games come out of book between moves 8-12 which is what I'm happy with.
Graham.