SCCT - Blitz Rating List 2015

[Sedat Canbaz]

Hello Chess Friends !)

I've just published my newest blitz rating list:
http://www.sedatcanbaz.com/chess/?page_id=1339

Note that,
I merged also the ponder on games, played in my previous 1 core blitz competitions
For examples: games from scratch; open-source ratings etc...too


Note also that,
The current competition is still active...I plan to work a little bit more over this project


Greetings,
Sedat

[Sedat Canbaz]

BTW, here is another new tournament with rapid time control:
http://www.sedatcanbaz.com/chess/?page_id=1326

[lkaufman]

Hello Chess Friends !)

I've just published my newest blitz rating list:
http://www.sedatcanbaz.com/chess/?page_id=1339

Note that,
I merged also the ponder on games, played in my previous 1 core blitz competitions
For examples: games from scratch; open-source ratings etc...too


Note also that,
The current competition is still active...I plan to work a little bit more over this project


Greetings,
Sedat

Thanks for running this. It is a pleasant surprise to see Komodo ahead of a quite recent version of Stockfish (just 3 weeks old) at a time control of just game/3', which is just about the worst time limit for us relative to SF. Also the Komodo version you used is older than the SF version, and I think we have more or less kept pace with SF since then. So this gives me reason to believe that once Stockfish 6 and Komodo 9 are released, Komodo may remain on top of the majority of the rating lists.

[beram]

Hello Chess Friends !)

I've just published my newest blitz rating list:
http://www.sedatcanbaz.com/chess/?page_id=1339

Note that,
I merged also the ponder on games, played in my previous 1 core blitz competitions
For examples: games from scratch; open-source ratings etc...too


Note also that,
The current competition is still active...I plan to work a little bit more over this project


Greetings,
Sedat

Thanks for running this. It is a pleasant surprise to see Komodo ahead of a quite recent version of Stockfish (just 3 weeks old) at a time control of just game/3', which is just about the worst time limit for us relative to SF. Also the Komodo version you used is older than the SF version, and I think we have more or less kept pace with SF since then. So this gives me reason to believe that once Stockfish 6 and Komodo 9 are released, Komodo may remain on top of the majority of the rating lists.

Than you havent yet seen his results at 15m 3s

..http://www.sedatcanbaz.com/chess/?page_id=1326

Code: Select all

SCCT Gladiators   2015

                          1                                       2                                       3                                       4                                       5                                       6                                       
1   Stockfish 221214 x64  20.0 - 19.026.0 - 13.026.0 - 13.021.5 - 15.529.0 - 9.0**      122.5/192
2   Komodo 1318 x64       19.0 - 20.020.5 - 17.523.0 - 15.022.5 - 15.526.5 - 12.5 **     111.5/192
3   Houdini 4 x64         13.0 - 26.017.5 - 20.517.5 - 21.520.0 - 18.028.0 - 10.0  **     96.0/192
4   Gull 3 x64            13.0 - 26.015.0 - 23.021.5 - 17.522.5 - 16.522.0 - 15.0   **    94.0/192
5   Fire 4 x64            15.5 - 21.515.5 - 22.518.0 - 20.016.5 - 22.520.0 - 19.0    **   85.5/191
6   Rybka 4.1 x64         9.0 - 29.012.5 - 26.510.0 - 28.015.0 - 22.019.0 - 20.0     **  65.5/191

[lkaufman]

Hello Chess Friends !)

I've just published my newest blitz rating list:
http://www.sedatcanbaz.com/chess/?page_id=1339

Note that,
I merged also the ponder on games, played in my previous 1 core blitz competitions
For examples: games from scratch; open-source ratings etc...too


Note also that,
The current competition is still active...I plan to work a little bit more over this project


Greetings,
Sedat

Thanks for running this. It is a pleasant surprise to see Komodo ahead of a quite recent version of Stockfish (just 3 weeks old) at a time control of just game/3', which is just about the worst time limit for us relative to SF. Also the Komodo version you used is older than the SF version, and I think we have more or less kept pace with SF since then. So this gives me reason to believe that once Stockfish 6 and Komodo 9 are released, Komodo may remain on top of the majority of the rating lists.

Than you havent yet seen his results at 15m 3s

..http://www.sedatcanbaz.com/chess/?page_id=1326

Code: Select all

SCCT Gladiators   2015

                          1                                       2                                       3                                       4                                       5                                       6                                       
1   Stockfish 221214 x64  20.0 - 19.026.0 - 13.026.0 - 13.021.5 - 15.529.0 - 9.0**      122.5/192
2   Komodo 1318 x64       19.0 - 20.020.5 - 17.523.0 - 15.022.5 - 15.526.5 - 12.5 **     111.5/192
3   Houdini 4 x64         13.0 - 26.017.5 - 20.517.5 - 21.520.0 - 18.028.0 - 10.0  **     96.0/192
4   Gull 3 x64            13.0 - 26.015.0 - 23.021.5 - 17.522.5 - 16.522.0 - 15.0   **    94.0/192
5   Fire 4 x64            15.5 - 21.515.5 - 22.518.0 - 20.016.5 - 22.520.0 - 19.0    **   85.5/191
6   Rybka 4.1 x64         9.0 - 29.012.5 - 26.510.0 - 28.015.0 - 22.019.0 - 20.0     **  65.5/191

      

Those are tiny numbers of games, about 1/8 of his blitz list, so of little concern so far.

[Sedat Canbaz]

Thanks for running this. It is a pleasant surprise to see Komodo ahead of a quite recent version of Stockfish (just 3 weeks old) at a time control of just game/3', which is just about the worst time limit for us relative to SF. Also the Komodo version you used is older than the SF version, and I think we have more or less kept pace with SF since then. So this gives me reason to believe that once Stockfish 6 and Komodo 9 are released, Komodo may remain on top of the majority of the rating lists.

Not at all...it's my pleasure

Yes, Komodo 1318 is real monster under these conditions...

And sometimes I am wondering,
Which one is stronger Stockfish or Komodo ?)
I think it all depends on used conditions...openings, time control, cores etc...

And as we see in my recent organized competitions...
We see sometimes SF sometimes Komodo to be the winner
In other words, it's quite exiting and interesting !)

Best,
Sedat

[Sedat Canbaz]

Code: Select all

SCCT Gladiators   2015

                          1                                       2                                       3                                       4                                       5                                       6                                       
1   Stockfish 221214 x64  20.0 - 19.026.0 - 13.026.0 - 13.021.5 - 15.529.0 - 9.0**      122.5/192
2   Komodo 1318 x64       19.0 - 20.020.5 - 17.523.0 - 15.022.5 - 15.526.5 - 12.5 **     111.5/192
3   Houdini 4 x64         13.0 - 26.017.5 - 20.517.5 - 21.520.0 - 18.028.0 - 10.0  **     96.0/192
4   Gull 3 x64            13.0 - 26.015.0 - 23.021.5 - 17.522.5 - 16.522.0 - 15.0   **    94.0/192
5   Fire 4 x64            15.5 - 21.515.5 - 22.518.0 - 20.016.5 - 22.520.0 - 19.0    **   85.5/191
6   Rybka 4.1 x64         9.0 - 29.012.5 - 26.510.0 - 28.015.0 - 22.019.0 - 20.0     **  65.5/191

191/192 games per player ?)

I think in the database there are 200 games per player (total 600 games)

Best,
Sedat

[lkaufman]

Thanks for running this. It is a pleasant surprise to see Komodo ahead of a quite recent version of Stockfish (just 3 weeks old) at a time control of just game/3', which is just about the worst time limit for us relative to SF. Also the Komodo version you used is older than the SF version, and I think we have more or less kept pace with SF since then. So this gives me reason to believe that once Stockfish 6 and Komodo 9 are released, Komodo may remain on top of the majority of the rating lists.

Not at all...it's my pleasure

Yes, Komodo 1318 is real monster under these conditions...

And sometimes I am wondering,
Which one is stronger Stockfish or Komodo ?)
I think it all depends on used conditions...openings, time control, cores etc...

And as we see in my recent organized competitions...
We see sometimes SF sometimes Komodo to be the winner
In other words, it's quite exiting and interesting !)

Best,
Sedat

Yes, you said it well, which one is stronger depends on the conditions. Komodo almost always comes out on top with 16 or more cores, but otherwise it depends on the things you mention plus on whether it is a match or a tournament. Komodo does better against other opponents than Stockfish (generally) but struggles a bit in direct matches. I have a theory about this but it is based on knowledge of the program so I can't explain it here.

[Sedat Canbaz]

Thanks for running this. It is a pleasant surprise to see Komodo ahead of a quite recent version of Stockfish (just 3 weeks old) at a time control of just game/3', which is just about the worst time limit for us relative to SF. Also the Komodo version you used is older than the SF version, and I think we have more or less kept pace with SF since then. So this gives me reason to believe that once Stockfish 6 and Komodo 9 are released, Komodo may remain on top of the majority of the rating lists.

Not at all...it's my pleasure

Yes, Komodo 1318 is real monster under these conditions...

And sometimes I am wondering,
Which one is stronger Stockfish or Komodo ?)
I think it all depends on used conditions...openings, time control, cores etc...

And as we see in my recent organized competitions...
We see sometimes SF sometimes Komodo to be the winner
In other words, it's quite exiting and interesting !)

Best,
Sedat

Yes, you said it well, which one is stronger depends on the conditions. Komodo almost always comes out on top with 16 or more cores, but otherwise it depends on the things you mention plus on whether it is a match or a tournament. Komodo does better against other opponents than Stockfish (generally) but struggles a bit in direct matches. I have a theory about this but it is based on knowledge of the program so I can't explain it here.

Agreed,
Probably Komodo MP is one of the rare engines, where with 16 or more cores is performing very good
It looks like mostly of the engines can not gain from the many cores...
And with a big possibility their Elo performance is affecting too...
That's why still I did not update my machines...

In other words,
Mostly of the passengers are still not ready to fly on the latest fast planes )!
And as a pilot, I don't wish to force those passengers in the risky areas )) !!


About in case of direct match: Komodo vs SF,
That's right...usually SF is doing better and here is an example (played at the current test):

Individual statistics:

1 Komodo 1318 x64 : 3340 1766 (+1226,=417,-123), 81.2 %

Stockfish 221214 x64 : 112 (+ 18,= 63,- 31), 44.2 %


Best,
Sedat

[Sedat Canbaz]

Last Updates:
Cheng4 0.38 x64 (+116 Elo over Cheng4 0.36 x64)
Stockfish 110115 x64 (+24 Elo over Stockfish 221214 x64)

Great improvement!! Well-done to both chess engine creators !

SCCT - All Versions:
http://www.sedatcanbaz.com/chess/?page_id=1339

SCCT - Top 20:
http://www.sedatcanbaz.com/chess/?page_id=1478

More details about the current competition:
- No double games
- No games up to 15 moves
- No games lost on time

In other words,
The rating database is quite clean... !)

But however,
I plan to publish separately the lost games on time (maybe they will be useful for the next engine releases)



Best,
Sedat