nTCEC Season 1 Stage 3 - 16 cores of chess fun

[Don]

Vitruvius 1.19 has now played 26 games without loss! Quite impressive, when Houdini has lost 3! What are statistics experts saying - new star or not?

I'm not an expert in statistics, but what matters is the full score, not how many games you did not lose.

Still, Vitruvius is doing impressively well and proving itself strong in this tournament.

[noctiferus]

As I posted before, I have some doubt about the significance of results from some openings.

Martin (Hi, Martin!) replied that now the openings are chosen by the community.
Anyway, IMH(patzer's)O, maybe that some openings could be still unbalanced toward one side (results 1-0, among equal strenght engines), so that, in order to have a credible idea about rankings of the stronger engines, we should wait, at least, for some opposite colours results.

My humble opinion...

[Don]

As I posted before, I have some doubt about the significance of results from some openings.

Martin (Hi, Martin!) replied that now the openings are chosen by the community.
Anyway, IMH(patzer's)O, maybe that some openings could be still unbalanced toward one side (results 1-0, among equal strenght engines), so that, in order to have a credible idea about rankings of the stronger engines, we should wait, at least, for some opposite colours results.

My humble opinion...

The openings in my opinion are of very minor consequence due to the fact that each player plays each side. Of course if a program is particular good at both side of some opening it might matter, but in general no big advantage is given by the openings to a particular program.

What can happen if one of the openings yield a big advantage for one side is that the results will tend to minimize the strength difference of one program over another. For example if the opening ends with a clear win for white, then you will likely get a win and a loss against each opponent you play against it. So it's not desirable to have either side end with a big advantage.

In general, the less the openings end with equal chances the more it favors the weaker programs. But I feel the impact is relatively minor.

[noctiferus]

"What can happen if one of the openings yield a big advantage for one side is that the results will tend to minimize the strength difference of one program over another. For example if the opening ends with a clear win for white, then you will likely get a win and a loss against each opponent you play against it. So it's not desirable to have either side end with a big advantage."

Exactly what I meant. You said it better...

[Don]

"What can happen if one of the openings yield a big advantage for one side is that the results will tend to minimize the strength difference of one program over another. For example if the opening ends with a clear win for white, then you will likely get a win and a loss against each opponent you play against it. So it's not desirable to have either side end with a big advantage."

Exactly what I meant. You said it better...

I apologize for repeating you.

What I really wanted to say that was different is that I don't think any of this really matters much.

[Martin Thoresen]

[moderation]
TCEC Tournament's thread moved to the tournaments and matches forum.

Oops, thanks Marcel. I didn't realize I posted it in the wrong forum.

I assume you are working full time now on Prédateur, clock is ticking.

[enhorning]

Halfway Standings:

Code: Select all

N Engine           Pts    SB R V K S H H Q C 

1 Rybka 4.1        5.5 16.75   = = 1 1 = 1 1 
2 Vitruvius 1.19   4.5 13.75 =   1 = = 0 1 1 
3 Komodo 4534      4.5 11.75 = 0   = = 1 1 1 
4 Stockfish 250313 4.5 11.25 0 = =   = 1 1 1 
5 Houdini 3        4.5 11.25 0 = = =   1 1 1 
6 Hiarcs 14        2.5  8.25 = 1 0 0 0   = = 
7 Quazar 0.4       1.5  1.75 0 0 0 0 0 =   1 
8 Chiron 1.5       0.5  1.25 0 0 0 0 0 = 0   

Seems to be a tight fight between 5 programs for 4 qualification slots! Komodo is doing impressively well, given that it is running on a single core, versus 16 cores for the other programs in the top-6.

[hgm]

In general, the less the openings end with equal chances the more it favors the weaker programs. But I feel the impact is relatively minor.

Actually it can be helpful to the strong program, if the level of play is high, and the imbalance is not too big. Obviously playing with Queen odds would make no sense, as it is a certain 1-1. But slightly below a Pawn can be good:

If the quality of play is so high that even the weaker player would usually not give up more than the draw margin against perfect play, most pairings would end in a pair of draws, a favorable result for the weaker player. By starting close to the win/draw boundary, only marginally stronger play is needed to salvage the draw when you get the short end, and secure the win when you play the advantaged side. Meaning most pairings would now end 1.5-0.5.

Only when the strength difference between the programs is so large that 25% is more than the weak player deservers does it become a disadvantage for the strong player.

[carldaman]

I find hard to understand how Chiron can crash so often that it had to be withdrawn. None of the testers, as far as I know, have reported any such problems. Chiron 1.5 running on my system also does not crash.

Could it be due to using too many cores and/or an incompatibility with the new 16-core machine that triggers a bug?

This is not a knock against Martin, though I think many of us would like to hear more of an explanation.

Thanks,
Carl

[Modern Times]

I find hard to understand how Chiron can crash so often that it had to be withdrawn. None of the testers, as far as I know, have reported any such problems. Chiron 1.5 running on my system also Carl

It does crash on just 4 cores, but not very often. I guess 16 cores makes the problem worse. I didn't know Martin had withdrawn it though.