invitation to a new deterministic no opening tournament

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

Uri Blass
Posts: 10788
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: invitation to a new deterministic no opening tournament

Post by Uri Blass »

RubiChess wrote: Wed Feb 21, 2024 4:43 pm What's that? You promised that "it [RubiChess] is too slow to participate in my tournament".
Unfortunately I did not find faster candidates to win against stockfish 3M nodes per move or berserk 5M nodes per move but maybe I did not make enough tries.

I may try later stockfish developement version but unfortunately it has no contempt and I will not be surprised if it get only draws even with 100M nodes per move.
Uri Blass
Posts: 10788
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: invitation to a new deterministic no opening tournament

Post by Uri Blass »

stockfish developement version could win with 100M nodes per move

Uri Blass
Posts: 10788
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: invitation to a new deterministic no opening tournament

Post by Uri Blass »

stockfish developement drew the second game with black.


Uri Blass
Posts: 10788
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: invitation to a new deterministic no opening tournament

Post by Uri Blass »

I added some engines.

Results so far:
standing

1)Stockfish15 10M 25.5
2)RubiChess contempt50 100M 25
3-4)stockfish16.1 30M 24.5
3-4)stockfish16.1 3M 24.5
5-6)stockfish16.1 1M 23.5
5-6)stockfish16.1 10M 23.5
7)Stockfish15 30M 22.5
8)Stockfish15 3M 21.5
9)Stockfish15 1M 21
10)stockfish16.1 300k 20
11)stockfish15 300k 18
12-13)stockfish15 100k 12.5
12-13)stockfish16.1 100k 12.5
14)berserk12 100k 10
15)stockfish16.1 30k 8.5
16)stockfish15 30k 6
17)stockfish16.1 10k 4.5
18)Stockfish15 10k 2.5

stockfish15 100M played 8 games against places 15-18 and got 2 out of 2 against 10k but only 1.5 out of 2 against 30k.

Interesting games
stockfish 100M failed to win against 30k
https://lichess.org/MkOxTlJe#188

Rubichess lost because of a bug by a stupid blunder.
It seems it had Nd5 in the hash for some reason when in the previous move it planned a different move based on the pv.
https://lichess.org/ayI4swFQ