SPCC: Testrun of Lc0 67741 finished

Discussion of computer chess matches and engine tournaments.

Moderators: hgm, Rebel, chrisw

User avatar
pohl4711
Posts: 2435
Joined: Sat Sep 03, 2011 7:25 am
Location: Berlin, Germany
Full name: Stefan Pohl

SPCC: Testrun of Lc0 67741 finished

Post by pohl4711 »

NN-testrun of Lc0 0.27.0 67741 finished

https://www.sp-cc.de

(Perhaps you have to clear your browsercache or reload the website)
Pedro
Posts: 26
Joined: Mon Oct 26, 2020 3:05 pm
Full name: Pedro

Re: SPCC: Testrun of Lc0 67741 finished

Post by Pedro »

Is this 3733 rating from Lc0 parallel with the ratings on the ab / NNUE engine list? Is it possible to compare?
User avatar
pohl4711
Posts: 2435
Joined: Sat Sep 03, 2011 7:25 am
Location: Berlin, Germany
Full name: Stefan Pohl

Re: SPCC: Testrun of Lc0 67741 finished

Post by pohl4711 »

Pedro wrote: Sat Feb 27, 2021 6:22 pm Is this 3733 rating from Lc0 parallel with the ratings on the ab / NNUE engine list? Is it possible to compare?
No. As I wrote on my website:
"500 NBSC-Advanced-Armageddon games each testrun (= a win for Black is 2 points for Black and a draw is a 1 point-win for Black). vs. Stockfish 200418 But mention, that the usage of my NBSC-Armageddon openings spreads the Elo-results around 2.25x wider, than using classical openings for testing(!), so with classical openings, you would need an errorbar of +/- 9 Elo for the same statistical quality of the results (= the rankings of Lc0 nets here). And for an errorbar of +/- 9 elo, you need around 3000 games, not 500, which means 6x more games (and 6x more PC-time)!!
Learn more about that revolution in computerchess in the "NBSC Armageddon openings"- section of my website."
Pedro
Posts: 26
Joined: Mon Oct 26, 2020 3:05 pm
Full name: Pedro

Re: SPCC: Testrun of Lc0 67741 finished

Post by Pedro »

pohl4711 wrote: Sat Feb 27, 2021 6:46 pm
Pedro wrote: Sat Feb 27, 2021 6:22 pm Is this 3733 rating from Lc0 parallel with the ratings on the ab / NNUE engine list? Is it possible to compare?
No. As I wrote on my website:
"500 NBSC-Advanced-Armageddon games each testrun (= a win for Black is 2 points for Black and a draw is a 1 point-win for Black). vs. Stockfish 200418 But mention, that the usage of my NBSC-Armageddon openings spreads the Elo-results around 2.25x wider, than using classical openings for testing(!), so with classical openings, you would need an errorbar of +/- 9 Elo for the same statistical quality of the results (= the rankings of Lc0 nets here). And for an errorbar of +/- 9 elo, you need around 3000 games, not 500, which means 6x more games (and 6x more PC-time)!!
Learn more about that revolution in computerchess in the "NBSC Armageddon openings"- section of my website."
Okay, so if we made a parallel with the list ab / NNUE, the elo of Lc0 would be 3641, right? But in this case, wouldn't it be weaker than the KomodoDragon,which has a 3647 elo rating on its list? But isn't Lc0 stronger than KDragon?