Berserk12 is slightly below SF14

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: Berserk12 is slightly below SF14

Post by Chessqueen »

ernest wrote: Sat Nov 11, 2023 4:36 am
Chessqueen wrote: Fri Nov 10, 2023 10:24 am
ernest wrote: Sun Nov 05, 2023 5:27 am
Chessqueen wrote: Sat Nov 04, 2023 10:52 am ...
Chances are that if you run it for another 1000 games up to 15000 games, Berserk 12 would probably come ahead by +3 :roll: :mrgreen: :roll:
Another sentence that shows you know nothing about statistics, poor Cheesequack !
When they are so close in rating the standard deviation could go either way, because data are close to the mean and not as reliable.
Your incompetence shows :
To go from 50.4 % (+3 Elo) at 15000 games to 49.6 % (-3 Elo) at 16000 games, Stockfish would need to be beaten 376 to 624 in the added 1000 games !!!
I am NOT going to continue this match to 14,000 or 15000 games, but there is a good probability of Berserk12 getting +3 if I do :roll:

Stockfish14 Vs Berserk12

Rank Engine Score Be St S-B
1 Berserk-12-x64-avx2 26.5/50 · ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· · ============1===========1=======1================= 622.75
2 Stockfish_14_x64_avx2 23.5/50 ============0===========0=======0================= · ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· · 622.75


50 games played / Tournament is finished

Tournament start: 2023.11.10, 06:28:46
Latest update: 2023.11.10, 17:54:44
Site/ Country: DESKTOP-4QNC0GS, United States
Level: Blitz 1/5
Hardware: Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz with 15.9 GB Memory
Operating system: Windows 10 Home Home Edition (Build 9200) 64 bit
PGN-File: Stockfish14 Vs Berserk12.pgn
Table created with: Arena 3.5.1
ernest
Posts: 2046
Joined: Wed Mar 08, 2006 8:30 pm

Re: Berserk12 is slightly below SF14

Post by ernest »

Chessqueen wrote: Sat Nov 11, 2023 9:34 am
ernest wrote: Sat Nov 11, 2023 4:36 am
Chessqueen wrote: Fri Nov 10, 2023 10:24 am
ernest wrote: Sun Nov 05, 2023 5:27 am
Chessqueen wrote: Sat Nov 04, 2023 10:52 am ...
Chances are that if you run it for another 1000 games up to 15000 games, Berserk 12 would probably come ahead by +3 :roll: :mrgreen: :roll:
Another sentence that shows you know nothing about statistics, poor Cheesequack !
When they are so close in rating the standard deviation could go either way, because data are close to the mean and not as reliable.
Your incompetence shows :
To go from 50.4 % (+3 Elo) at 15000 games to 49.6 % (-3 Elo) at 16000 games, Stockfish would need to be beaten 376 to 624 in the added 1000 games !!!
I am NOT going to continue this match to 14,000 or 15000 games, but there is a good probability of Berserk12 getting +3 if I do :roll:

Stockfish14 Vs Berserk12

Rank Engine Score Be St S-B
1 Berserk-12-x64-avx2 26.5/50 · ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· · ============1===========1=======1================= 622.75
2 Stockfish_14_x64_avx2 23.5/50 ============0===========0=======0================= · ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· · 622.75


50 games played
I was answering your stupid sentence
"Chances are that if you run it for another 1000 games up to 15000 games, Berserk 12 would probably come ahead by +3 :roll: :mrgreen: :roll:"
...not your irrelevant own match !
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: Berserk12 is slightly below SF14

Post by Chessqueen »

ernest wrote: Sun Nov 12, 2023 2:24 am
Chessqueen wrote: Sat Nov 11, 2023 9:34 am
ernest wrote: Sat Nov 11, 2023 4:36 am
Chessqueen wrote: Fri Nov 10, 2023 10:24 am
ernest wrote: Sun Nov 05, 2023 5:27 am
Chessqueen wrote: Sat Nov 04, 2023 10:52 am ...
Chances are that if you run it for another 1000 games up to 15000 games, Berserk 12 would probably come ahead by +3 :roll: :mrgreen: :roll:
Another sentence that shows you know nothing about statistics, poor Cheesequack !
When they are so close in rating the standard deviation could go either way, because data are close to the mean and not as reliable.
Your incompetence shows :
To go from 50.4 % (+3 Elo) at 15000 games to 49.6 % (-3 Elo) at 16000 games, Stockfish would need to be beaten 376 to 624 in the added 1000 games !!!
I am NOT going to continue this match to 14,000 or 15000 games, but there is a good probability of Berserk12 getting +3 if I do :roll:

Stockfish14 Vs Berserk12

Rank Engine Score Be St S-B
1 Berserk-12-x64-avx2 26.5/50 · ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· · ============1===========1=======1================= 622.75
2 Stockfish_14_x64_avx2 23.5/50 ============0===========0=======0================= · ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· ·· · 622.75


50 games played
I was answering your stupid sentence
"Chances are that if you run it for another 1000 games up to 15000 games, Berserk 12 would probably come ahead by +3 :roll: :mrgreen: :roll:"
...not your irrelevant own match !
50 games are NOT irrelevant otherwise the World Chess Championship between human needs to be at least 1000 games, instead of wasting y0ur time answering my post here why you do NOT try to beat at least Mr. Oscar Bot and satisfy your Chess Ego ===>https://www.youtube.com/watch?v=V-f_t84wk6w
Ciekce
Posts: 190
Joined: Sun Oct 30, 2022 5:26 pm
Full name: Conor Anstey

Re: Berserk12 is slightly below SF14

Post by Ciekce »

Chessqueen wrote: Sun Nov 12, 2023 10:35 am 50 games are NOT irrelevant otherwise the World Chess Championship between human needs to be at least 1000 games
Yes, the world championship is a very small and non-statistically significant sample. Your lack of understanding of statistics is showing again.
dkappe
Posts: 1632
Joined: Tue Aug 21, 2018 7:52 pm
Full name: Dietrich Kappe

Re: Berserk12 is slightly below SF14

Post by dkappe »

Ciekce wrote: Sun Nov 12, 2023 8:50 pm
Chessqueen wrote: Sun Nov 12, 2023 10:35 am 50 games are NOT irrelevant otherwise the World Chess Championship between human needs to be at least 1000 games
Yes, the world championship is a very small and non-statistically significant sample. Your lack of understanding of statistics is showing again.
Don’t feed the troll. The best situation is ChessQueen responding to themselves for a dozen messages and nobody else paying attention.
Fat Titz by Stockfish, the engine with the bodaciously big net. Remember: size matters. If you want to learn more about this engine just google for "Fat Titz".
Ciekce
Posts: 190
Joined: Sun Oct 30, 2022 5:26 pm
Full name: Conor Anstey

Re: Berserk12 is slightly below SF14

Post by Ciekce »

dkappe wrote: Mon Nov 13, 2023 4:08 am Don’t feed the troll. The best situation is ChessQueen responding to themselves for a dozen messages and nobody else paying attention.
fair, I'm too easy to bait
Chessqueen
Posts: 5685
Joined: Wed Sep 05, 2018 2:16 am
Location: Moving
Full name: Jorge Picado

Re: Berserk12 is slightly below SF14

Post by Chessqueen »

dkappe wrote: Mon Nov 13, 2023 4:08 am
Ciekce wrote: Sun Nov 12, 2023 8:50 pm
Chessqueen wrote: Sun Nov 12, 2023 10:35 am 50 games are NOT irrelevant otherwise the World Chess Championship between human needs to be at least 1000 games
Yes, the world championship is a very small and non-statistically significant sample. Your lack of understanding of statistics is showing again.
Don’t feed the troll. The best situation is ChessQueen responding to themselves for a dozen messages and nobody else paying attention.
You are correct Do Not feed the Troll, We all know that he enjoys doing that, I do NOT know what he gets from doing it over and over again, but my advice is NOT to respond to any of his teasing posts, so Now you all know what to do; Good Night. Now My good side or my Saint side told you what to do. :mrgreen: