SPCC: Testrun of asmFish 170310 finished

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

User avatar
pohl4711
Posts: 2950
Joined: Sat Sep 03, 2011 7:25 am
Location: Berlin, Germany
Full name: Stefan Pohl

SPCC: Testrun of asmFish 170310 finished

Post by pohl4711 »

Testrun of asmFish 170310 finished.

http://spcc.beepworld.de

(Perhaps you have to clear your browsercache or reload the website)
Top
leavenfish
Posts: 282
Joined: Mon Sep 02, 2013 8:23 am

Re: SPCC: Testrun of asmFish 170310 finished

Post by leavenfish »

So, I don't understand...how can BrainFish be above the others? I mean, isn't it simply the standard Stockfish engine augmented by a special playing book...but these tests all use the same book. Confused... :?
Top
Dann Corbit
Posts: 12870
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: SPCC: Testrun of asmFish 170310 finished

Post by Dann Corbit »

leavenfish wrote:So, I don't understand...how can BrainFish be above the others? I mean, isn't it simply the standard Stockfish engine augmented by a special playing book...but these tests all use the same book. Confused... :?
All except brainfish, which gets to use the special book:
Cerebellum_Light.bin
which has 43 megabytes of mini-maxed chess analysis.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
Top
User avatar
pohl4711
Posts: 2950
Joined: Sat Sep 03, 2011 7:25 am
Location: Berlin, Germany
Full name: Stefan Pohl

Re: SPCC: Testrun of asmFish 170310 finished

Post by pohl4711 »

Dann Corbit wrote:
leavenfish wrote:So, I don't understand...how can BrainFish be above the others? I mean, isn't it simply the standard Stockfish engine augmented by a special playing book...but these tests all use the same book. Confused... :?
All except brainfish, which gets to use the special book:
Cerebellum_Light.bin
which has 43 megabytes of mini-maxed chess analysis.
Correct. The Brainfish-Testruns were done with the Cerebellum-Library. Thomas Zipproth and me wanted to measure the Elo-gain of Cerebellum.

Stefan (SPCC)
Top
User avatar
Leto
Posts: 2152
Joined: Thu May 04, 2006 3:40 am
Location: Dune

Re: SPCC: Testrun of asmFish 170310 finished

Post by Leto »

Stockfish 161212 x64 : 3407 7 7 7000 76.4 % 3187 35.8 %
Stockfish 170318 x64 : 3402 7 7 7000 74.4 % 3203 36.3 %

Three months no progress?
Top
User avatar
Eelco de Groot
Posts: 4724
Joined: Sun Mar 12, 2006 2:40 am
Full name:   Eelco de Groot

Re: SPCC: Testrun of asmFish 170310 finished

Post by Eelco de Groot »

Leto wrote:Stockfish 161212 x64 : 3407 7 7 7000 76.4 % 3187 35.8 %
Stockfish 170318 x64 : 3402 7 7 7000 74.4 % 3203 36.3 %

Three months no progress?
Really weird, the line is now seemingly completely flat, both of Stockfish and AsmFish, distance between them is constant. I believe Komodo 10.4 is a new opponent but that also does not make any bumps in the line that I can see.
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
Top
User avatar
pohl4711
Posts: 2950
Joined: Sat Sep 03, 2011 7:25 am
Location: Berlin, Germany
Full name: Stefan Pohl

Re: SPCC: Testrun of asmFish 170310 finished

Post by pohl4711 »

Leto wrote:Stockfish 161212 x64 : 3407 7 7 7000 76.4 % 3187 35.8 %
Stockfish 170318 x64 : 3402 7 7 7000 74.4 % 3203 36.3 %

Three months no progress?
Sadly, thats true. There were a lot of functional patches applied to the Stockfish-code in the last three months, but no measureable Elo-gain against non-Stockfish opponents in my testruns.
Top

Return to “Computer Chess Club: Tournaments and Matches”