Stockfish 4 running for the IPON

[beram]

.

... My own testing agrees extremely well with Ingo's, in that I show that although Komodo 5.1 still has a lead over SF4 in direct play, it has shrunk from a large lead to a quite small one.... Obviously we need to improve Komodo or SF will soon pass us.

Well Larry I very much doubt that
or perhaps you test Komodo 51 with contempt = 0
or the SSE42
or something else you surely will come up with

Sorry, I can't understand your point. You find it strange that my testing agrees with Ingo's?? Are you saying that his result was somehow wrong? Too high for SF or too low? Why?

I mean to say that IMO SF4 MP(!) already has passed Komodo 51MP
and not only my testings confirm this

[lkaufman]

.

... My own testing agrees extremely well with Ingo's, in that I show that although Komodo 5.1 still has a lead over SF4 in direct play, it has shrunk from a large lead to a quite small one.... Obviously we need to improve Komodo or SF will soon pass us.

Well Larry I very much doubt that
or perhaps you test Komodo 51 with contempt = 0
or the SSE42
or something else you surely will come up with

Sorry, I can't understand your point. You find it strange that my testing agrees with Ingo's?? Are you saying that his result was somehow wrong? Too high for SF or too low? Why?

I mean to say that IMO SF4 MP(!) already has passed Komodo 51MP
and not only my testings confirm this

OK, I was talking about testing on one core. Now that all the top engines have MP and get roughly similar elo gains from it, testing on single core should produce similar results to testing on MP if the time limits are adjusted to be of comparable quality. Maybe Stockfish gains slightly more from MP than other programs, I don't know. Single core testing allows for greater sample size for the same level.

[ernest]

Stockfish 4 run is finished and online:

Hi Ingo,

Maybe this is irrelevant when using only 1 thread per engine,
but did you leave for Stockfish 4 the default Idle Threads Sleep false?

[IWB]

Stockfish 4 run is finished and online:

Hi Ingo,

Maybe this is irrelevant when using only 1 thread per engine,
but did you leave for Stockfish 4 the default Idle Threads Sleep false?

As usuall I played with default settings (as 99% of all users) and that is "Idle Threads Sleep" unchecked.

Whats that good for at all?

Bye
Ingo

[zullil]

Note that contempt is only one explanation for doing better against weak opponents.

I did not watch the games so I do not know but I can think about an alternative explanation and maybe stockfish is relatively worse in converting advantages to win regardless of contempt and I saw cases when stockfish evaluated some drawn endgame as winning.

for example see the following position and give stockfish to search

Stockfish is happy to get it against weak opponents instead of winning against them.

[d]7k/8/8/8/8/7P/6K1/7B w - - 0 1

Fixed by the latest commit:

Code: Select all

Searching: 7k/8/8/8/8/7P/6K1/7B w - - 0 1
infinite: 0 ponder: 0 time: 0 increment: 0 moves to go: 0
 1   +0.00   00:00      14  h4 Kg7 
 2   +0.00   00:00      57  h4 Kg7 h5 Kh6 
 3   +0.00   00:00     158  h4 Kg7 h5 Kh6 Kf1 
 4   +0.00   00:00     365  h4 Kg7 h5 Kh6 Kf1 Kxh5 
 5   +0.00   00:00     814  h4 Kg7 h5 Kh6 Kf1 Kxh5 
 6   +0.00   00:00    1323  h4 Kg7 h5 Kh6 Kf1 Kxh5 
 7   +0.00   00:00    2196  h4 Kg7 h5 Kh6 Kf1 Kxh5 
 8   +0.00   00:00    3955  h4 Kg7 h5 Kh6 Kf1 Kxh5 
 9   +0.00   00:00    6725  h4 Kg7 h5 Kh6 Kf1 Kxh5 
10   +0.00   00:00   11268  h4 Kg7 h5 Kh6 Kf1 Kxh5 
11   +0.00   00:00   17880  h4 Kg7 h5 Kh6 Kf1 Kxh5 
12   +0.00   00:00   26765  h4 Kg7 h5 Kh6 Kf1 Kxh5 
13   +0.00   00:00   39377  h4 Kg7 h5 Kh6 Kf1 Kxh5 
14   +0.00   00:00   55579  h4 Kg7 h5 Kh6 Kf1 Kxh5 
15   +0.00   00:00   77523  h4 Kg7 h5 Kh6 Kf1 Kxh5 
16   +0.00   00:00  104201  h4 Kg7 h5 Kh6 Kf1 Kxh5 
17   +0.00   00:00  133933  h4 Kg7 h5 Kh6 Kf1 Kxh5 
18   +0.00   00:00  176885  h4 Kg7 h5 Kh6 Kf1 Kxh5 
19   +0.00   00:00  230759  h4 Kg7 h5 Kh6 Kf1 Kxh5 
20   +0.00   00:00  290505  h4 Kg7 h5 Kh6 Kf1 Kxh5 
21   +0.00   00:00  372797  h4 Kg7 h5 Kh6 Kf1 Kxh5 
22   +0.00   00:00  484163  h4 Kg7 h5 Kh6 Kf1 Kxh5 
23   +0.00   00:00  609564  h4 Kg7 h5 Kh6 Kf1 Kxh5 
24   +0.00   00:00  774614  h4 Kg7 h5 Kh6 Kf1 Kxh5 
25   +0.00   00:00  976212  h4 Kg7 h5 Kh6 Kf1 Kxh5 
26   +0.00   00:00   1187K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
27   +0.00   00:00   1442K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
28   +0.00   00:00   1909K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
29   +0.00   00:00   2404K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
30   +0.00   00:00   2882K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
31   +0.00   00:00   3173K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
32   +0.00   00:00   3822K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
33   +0.00   00:00   4104K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
34   +0.00   00:01   4834K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
35   +0.00   00:01   5637K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
36   +0.00   00:01   6084K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
37   +0.00   00:01   6714K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
38   +0.00   00:01   7319K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
39   +0.00   00:01   8161K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
40   +0.00   00:02   9010K  h4 Kg7 h5 Kh6 Kf1 Kxh5 
Nodes: 9010090
Nodes/second: 4462649
Best move: h4
Ponder move: Kg7