Stockfish "111026" simply has 10% faster NPS in my PC than 2.1.1 JA, so that should give +10 ELO automatically. And possible there is also additional improvements too.
Jouni
Finally a better Stockfish?
Moderators: hgm, Rebel, chrisw
-
- Posts: 3304
- Joined: Wed Mar 08, 2006 8:15 pm
Re: Finally a better Stockfish?
I played some fast games. 111026 scores +20 points against 2.1.1 after 200 games, but vs. Houdini it was actually worse.. OK 2000 games is there minimum
Jouni
Jouni
-
- Posts: 766
- Joined: Sun Oct 16, 2011 11:25 am
Re: Finally a better Stockfish?
Hi, i'm watching on playchess Stockfish 111030, but i cannot find a link..Jouni wrote:I played some fast games. 111026 scores +20 points against 2.1.1 after 200 games, but vs. Houdini it was actually worse.. OK 2000 games is there minimum
Jouni
Regards
MM
-
- Posts: 1972
- Joined: Wed Jul 13, 2011 9:04 pm
- Location: Madrid, Spain.
Re: Finally a better Stockfish?
Hello:
https://github.com/mcostalba/Stockfish/ ... e44031d2b5
There is no functional change, according to Costaba. I also have not found a link, but it can be compiled (not by me, of course, because I do not know).
Regards from Spain.
Ajedrecista.
https://github.com/mcostalba/StockfishMM wrote:Hi, i'm watching on playchess Stockfish 111030, but i cannot find a link..Jouni wrote:I played some fast games. 111026 scores +20 points against 2.1.1 after 200 games, but vs. Houdini it was actually worse.. OK 2000 games is there minimum
Jouni
Regards
https://github.com/mcostalba/Stockfish/ ... e44031d2b5
Code: Select all
Sync do_move() and undo_move()
It is not possible to unify due to the fact that the
sequence steps are reversed. What we can do is to try
to sync comments and code as much as we can to easy
reading and documentation.
No functional change.
Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Regards from Spain.
Ajedrecista.
-
- Posts: 4571
- Joined: Sun Mar 12, 2006 2:40 am
- Full name:
Re: Finally a better Stockfish?
I believe Marco is making some code clean-ups and simplifying things where code is no longer needed. Mainly in the weekends He always adds if there is a functional change. If there is no functional change, the output of the new engine should be 100.000% identical to the previous version. It is of course fine if this is tested now that the changes can be seen right away. But I fear a lot of people are now expecting that this are somehow stronger versions of Stockfish. That is not very likely. Everybody can read for themselves what the changes are, if you go halfway the main page and click 'history' on the right I admit I also did not find this out right away. This is the list of Marco's commits:
https://github.com/mcostalba/Stockfish/commits/master/
Click on the +- buttons after every commit to read what the change entails and if there is a functional change being made to the master branch. The functional changes are I believe mainly from CLOP tuning experiments, Remi's tool for simulated annealing, and that is limited for the moment, as far as I could see, to some changes in mobility
Regards, Eelco
https://github.com/mcostalba/Stockfish/commits/master/
Click on the +- buttons after every commit to read what the change entails and if there is a functional change being made to the master branch. The functional changes are I believe mainly from CLOP tuning experiments, Remi's tool for simulated annealing, and that is limited for the moment, as far as I could see, to some changes in mobility
Regards, Eelco
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
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan
-
- Posts: 2684
- Joined: Sat Jun 14, 2008 9:17 pm
Re: Finally a better Stockfish?
Yes, correctEelco de Groot wrote:I believe Marco is making some code clean-ups and simplifying things where code is no longer needed. Mainly in the weekends
It is also correct that current version should be around +20 ELO (on SMP) against 2.1.1 but not because of latest changes. It is due to some work, done mainly by Joona, some month ago. Of course this is not enough to do a new release and anyhow I am not interested at the moment. I'd just experimenting a bit with CLOP and doing small things in the weekends...
-
- Posts: 134
- Joined: Mon May 16, 2011 6:58 pm
- Location: Denmark
Re: Finally a better Stockfish?
So it will be able to beat Critter?
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: Finally a better Stockfish?
Code: Select all
Program Score % Elo + - Draws
1 Stockfish 111026 64bit : 524.5/1000 52.4 3209 16 16 46.1 %
2 Stockfish 2.1.1 JA 64bit : 475.5/1000 47.5 3191 16 16 46.1 %
Kai
-
- Posts: 1972
- Joined: Wed Jul 13, 2011 9:04 pm
- Location: Madrid, Spain.
Re: Finally a better Stockfish?
Hello again:
I have just found a compilation of Stockfish 111030:
http://www.chess2u.com/t4410p45-stockfish-01#27006
http://www.mediafire.com/?3m6bkanv643gbef
(Post #51). Enjoy!
Regards from Spain.
Ajedrecista.
I have just found a compilation of Stockfish 111030:
http://www.chess2u.com/t4410p45-stockfish-01#27006
http://www.mediafire.com/?3m6bkanv643gbef
(Post #51). Enjoy!
Regards from Spain.
Ajedrecista.
-
- Posts: 766
- Joined: Sun Oct 16, 2011 11:25 am
Re: Finally a better Stockfish?
Hi, thank you.Ajedrecista wrote:Hello again:
I have just found a compilation of Stockfish 111030:
http://www.chess2u.com/t4410p45-stockfish-01#27006
http://www.mediafire.com/?3m6bkanv643gbef
(Post #51). Enjoy!
Regards from Spain.
Ajedrecista.
It is much slower than 2.1.1
MM