WCCC 2013 in Yokohama

[hgm]

No, it is common sense.

What is just an opinion (or an idiocy, depending on who you ask it) is that a non-existing engine would crush a 3000+ engine running on the fastest hardware available at long TC in 25 moves.

[overlord]

If you let play blitz match between Houdini and Stockfish (for example), you will see that most games that Stockfish lost are caused by the fact that Stockfish missed Houdini next move and due to limited time it moves some blunder at low depths. Stockfish is very strong in particular positions and at deep depths, but Houdini search algorithm is totally different so even at very low depths probability of terrible blunder is not so high.

[overlord]

Crushing means that the game is already decided...I am not writing about mate in 25 moves. Only small kids are playing until the end.

[hgm]

So you are saying now that Houdini wins because it plays better moves after a ponder miss. Which has absolutely nothing to do with the ponder-hit rate.

[hgm]

Crushing means that the game is already decided...I am not writing about mate in 25 moves. Only small kids are playing until the end.

It still doesn't seem to register... How would an entity that does not exist even do a single move? The game would be totally decided within 1 move, alright. By time forfeit. And not in favor of Houdini Android...

[overlord]

Houdini wins mainly because it better then the others. Houdini benefits in both cases, if Houdini hits opponents move (is it proable that it is good move) it is good for him because he analysed it deeply on opponents time. If Houdini miss opponents move it is also goodd for Houdini becuase there is high chance that oppoents move was simply weak so houdini doesn´t need a lot of time to punish it. That is the main goal. Simply said stronger engine benefits from ponder on more then weaker one...

[hgm]

That is nonsense, because Houdini would not take less time to think on a move after a ponder miss. And it obviously cannot benefit from something it doesn't do.

It would also be an collosal mistake (gravely weakening any engine) to get careless and think shorter just because you pedicted the opponent's move wrong. Even if you are stronger. Because fingerprinting tests have shown that move preferences depend more on strategic (evaluation) difference between engines, than on tactical ability (search time). Moves that are suboptimal from a strategic point of view in general are not any less tactically dangerous.

[IanO]

[The] problem is it ought to be discontinued. It has turned into nothing more than a sick joke. Junior won last time around. In my beta testing for Don, at reasonable time controls- his WEAKEST beta that I tested in 30 game matches, in just wins and losses, beat Junior 13.3 and Deep Fritz 13 by a combined score of 40-0! Neither engine even got a win in 60 games- and I am still amazed they figured out how to draw 20 games between them. Guess they got lucky.

The best analogy I can come up with is that it is time for the World Cup championship game- and both sides come down with the flu. So instead of cancelling it- they run in 2 junior high school soccer teams to play each other for the World Cup.Kids 10 and 11 years old, and still have the gall to insult people's intelligence by attempting to drum up interest for it. And the winner advertises itself as the best team in existence in the world. Which do you do- laugh at them or cry.


gts

Not the best analogy. Teams getting sick implies that they wanted to participate in the first place. A better analogy is Olympic Baseball as the world championship (ICGA) whereas all the best teams actually play in the US leagues (rating lists). The problem is that the ICGA event is so archaic, inconvenient, and expensive, that no modern authors want to participate. They were born in the internet age and are used to the convenience of rating lists and online tournaments.

The non-plagiarism rules also put a damper on new entrants. Half of the top entrants in rating lists would probably be disputed; Komodo and Stockfish are the exception rather than the rule.

I agree that this event should drop the World Championship title, since it obviously isn't, and simply be the Computer Olympiad chess events.

Ian

[fern]

haha
I think the same.

Playing superconnie regards
Fern

[hgm]

I agree that this event should drop the World Championship title, since it obviously isn't, and simply be the Computer Olympiad chess events.

I see no point in that. Every self-respecting competitive sport needs a world title. This seems the best format, and year after year it attracts many of the best Chess programmers in the world. Alas, not all, but you can't have everything. If you would only do things when they could be done perfectly, not much would ever get done...

Fact is that when an on-line format was proposed, exactly one participant showed up. (And he was not even an author.) The current format apparently does not suffer from that at all.