Tech (average nodes, depths, time/m per move, others per game), counted for computing moves only, ignored moves with zero nodes: # name nodes/m NPS depth/m time/m moves time 1. Stockfish 14 5m+5s 263911K 29566552 52.6 8.9 65.0 580.2 2. Stockfish 14 60m+60s 3341118K 30431351 67.2 109.8 64.9 7127.3 all --- 1759304K 30366252 59.9 59.3 65.0 3853.8
The question is if other engines with 60+60 or slower time control can win against stockfish14 with 5+5 You can try stockfish11 with 60+60 against stockfish14 with 5+5 or maybe some engine that does not use NNUE or maybe slower time conrol than 60+60 for a different engine.
I tried a match of 10 games in FRC of stockfish11 20+10 against one stockfish developement version from 9.10 2+1 and the result was 1-1 and 8 draws but stockfish developement version seemed stupid in the only game that it lost with white and showed 0.00 for a long time in lost position.
Note that my hardware is slower than your hardware and I gave 7 threads for every engine.
summary of the match Result: ------------------------------------------------------------------------------------ # name games wins% draws% losses% score elo + - 1. Stockfish 091021 10 10.0 80.0 10.0 5 1 72 72 2. Stockfish 11 64 POPCNT 10 10.0 80.0 10.0 5 -1 72 72
Cross table: ------------------------------------------------------------------------------------ # name score games 1 2 1. Stockfish 091021 5 10 x =1=0====== 2. Stockfish 11 64 POPCNT 5 10 =0=1====== x
only when stockfish developement version played 32.Rc1 it saw it is losing inspite of the fact Bishop a2 is out of the game(of course with more time it can see it earlier)
some analysis by one stockfish developement version for the position after 30...Bf8 with 7 cores(see the 0.00 at depth 20)
Yes, but not totally unexpected. As are scaling data predicted this results with Stockfish. It is just more dramatic to see it in real games, and at such an extreme level.
As this is like a 16 core system playing 192 cores system at 5m+5s running Stockfish 14.
Last edited by mwyoung on Fri Oct 22, 2021 9:16 pm; edited 2 times in total
Admin Admin
Posts : 2528 Join date : 2020-11-17 Location : Netherlands
Subject: Re: Stockfish 14 60m+60s vs Stockfish 14 5m+5s Fri Oct 22, 2021 8:44 pm
Curious how Dragon will do with your book.
You already made your point with SF14
mwyoung
Posts : 880 Join date : 2020-11-25 Location : USA
Subject: Re: Stockfish 14 60m+60s vs Stockfish 14 5m+5s Fri Oct 22, 2021 8:47 pm
Admin wrote:
Curious how Dragon will do with your book.
You already made your point with SF14
OK I will start Dragon 2.5
Admin likes this post
Uri Blass
Posts : 206 Join date : 2020-11-28
Subject: Re: Stockfish 14 60m+60s vs Stockfish 14 5m+5s Fri Oct 22, 2021 8:50 pm
Admin wrote:
Curious how Dragon will do with your book.
You already made your point with SF14
I would like to see if Stockfish11 that does not use NNUE with 60+60 can get more wins against Stockfish14 5+5 with the same number of games.
If there is some better non NNUE engine than Stockfish11 that is not stockfish developement version earlier to stockfish12 when the difference to stockfish11 is small(I do not know) then maybe it is better to use it.
mwyoung
Posts : 880 Join date : 2020-11-25 Location : USA
Subject: Re: Stockfish 14 60m+60s vs Stockfish 14 5m+5s Fri Oct 22, 2021 9:26 pm
End of test.
Code:
Result: ---------------------------------------------------------------------------------- # name games wins draws losses score elo + - 1. Stockfish 14 60m+60s 23 1 22 0 12 4 48 47 2. Stockfish 14 5m+5s 23 0 22 1 11 -4 47 48
Cross table: ---------------------------------------------------------------------------------- # name score games 1 2 1. Stockfish 14 60m+60s 12 23 x ==================1====----------------------------------------------------------------------------- 2. Stockfish 14 5m+5s 11 23 ==================0====----------------------------------------------------------------------------- x
Home 
