Andscacs - New version 0.921 with source

[Henk]

Looks to me Andscacs is playing too many draws. Intention should be to destroy your opponent completely otherwise better not play chess. Better lose than going for a draw.

That's a matter of opinion.

I am still able to think about something more friendly than capturing a king. Or what do you mean. Is it about Andscacs playing not too many draws or the intention of playing chess.

[ernest]

As the bug happened in only one move situation,

...
If this is indeed what happened, then the solution is to reset the various thread-specific data structures in the main thread before starting the actual search.

Bravo, Ron !
Seems the craziest bugs can be corrected through logical thinking...

[Gusev]

Thank you!! The source code release makes Andscacs eligible for participation in the next season of FOSCEC. Moreover, CCRL should change the engine name color from green to orange!

[tpoppins]

[Guenther]

Formatted table as real data. (pics vanish and cannot be used for anything)

CCRL 40/40 Rating List - Custom engine selection (November 18, 2017)

Code: Select all

Name                            Rating  Err1    Err2    Score   AvgOpp. Draws   Games
Andscacs 0.921 64-bit 4CPU      3274    +67     -71     26.6%   +138.8  50.0%   64 
Andscacs 0.91 64-bit 4CPU       3239    +22     -22     42.5%   +47.6   59.2%   596
Andscacs 0.89 64-bit 4CPU       3232    +21     -21     45.9%   +28.8   55.0%   662
Andscacs 0.88 64-bit 4CPU       3228    +37     -37     33.2%   +100.4  53.5%   217
Andscacs 0.90 64-bit 4CPU       3216    +24     -24     50.3%   +1.9    55.1%   515
Andscacs 0.872 64-bit 4CPU      3209    +24     -24     51.3%   -5.0    58.8%   476
Andscacs 0.92 64-bit            3161    +15     -15     53.1%   -15.7   52.6%   1344
Andscacs 0.86 64-bit 4CPU       3154    +22     -22     40.0%   +64.1   54.6%   639
Andscacs 0.85 64-bit 4CPU       3142    +25     -25     43.9%   +42.7   53.1%   467
Andscacs 0.91 64-bit            3140    +16     -16     44.0%   +37.7   56.4%   1189
Andscacs 0.921 64-bit           3138    +31     -31     42.7%   +49.4   50.2%   307
Andscacs 0.90 64-bit            3127    +17     -17     48.4%   +13.0   53.8%   1048
Andscacs 0.89 64-bit            3119    +17     -17     43.4%   +44.2   54.4%   1029
Andscacs 0.88 64-bit            3114    +15     -15     43.6%   +41.4   54.1%   1266
Andscacs 0.83 64-bit 4CPU       3113    +29     -29     46.6%   +18.5   51.2%   338
Andscacs 0.872 64-bit           3107    +14     -14     43.8%   +41.1   52.1%   1603
Andscacs 0.82 64-bit 4CPU       3090    +29     -29     50.6%   -3.2    55.2%   330
Andscacs 0.86 64-bit            3090    +21     -21     44.7%   +34.9   51.3%   678
Andscacs 0.81 64-bit 4CPU       3085    +31     -32     48.7%   +5.3    49.3%   300
Andscacs 0.85 64-bit            3061    +18     -18     44.9%   +33.6   48.3%   918
Andscacs 0.84 64-bit            3041    +26     -26     45.1%   +32.3   46.8%   470
Andscacs 0.80 64-bit 4CPU       3030    +31     -31     50.2%   -1.1    46.7%   317
Andscacs 0.83 64-bit            3014    +30     -30     47.5%   +14.7   50.8%   317
Andscacs 0.82 64-bit            3010    +24     -24     47.8%   +12.4   50.1%   509
Andscacs 0.81 64-bit            3003    +26     -26     55.1%   -31.5   47.3%   442

[tpoppins]

Günther, the point of my post was the color -- something your carefully formatted text table fails to convey. ;)

[kasinp]

Günther, the point of my post was the color -- something your carefully formatted text table fails to convey.

Here is what I see:

a) 4-CPU direct comparison between 0.921 and 0.92 is not available (ver. 0.92 not having been tested with 4CPUs)
b) However, a comparison between single CPU performance indicates that ver. 0.921 (based on some 300 games) is 23 ELO *below* that of 0.92.
c) Therefore the fact that 0.921 is leading the charts for 4CPUs is misleading, and probably only because 0.92 was not tested with 4CPUs.

This would support the hypothesis that 0.921 may have have introduced an issue, irrespective of the illegal move played in one of the TCEC games.

PK

[cdani]

... irrespective of the illegal move played in one of the TCEC games.

The bug of the illegal move was there from very old versions. Only that seems that a lot of cores and bad luck are necessary to manifest it.

[kasinp]

... irrespective of the illegal move played in one of the TCEC games.

The bug of the illegal move was there from very old versions. Only that seems that a lot of cores and bad luck are necessary to manifest it.

Got it, thank you for the clarification. I suppose this still leaves the question of a direct comparison between 0.92 and 0.921, for which I suppose 300+ games is probably still a small sample size.

[Guenther]

Guenther, the point of my post was the color -- something your carefully formatted text table fails to convey. ;)

Sorry, Thomas :)
A pity that <code> doesn't allow color too.