Houdini 3 vs ClusterRybka Game 4 ...

[Lion]

Ok so we can say the cluster lost the match.

[Leto]

Ok so we can say the cluster lost the match.

Yes, it's over. Would they still play the remaining games?

[S.Taylor]

Yet another time, this game has shown that the Rybka Cluster evaluation is always one step back from the one of Houdini.

Simply speaking Houdini sees things much earlier than Rybka can, and it even sees things that Rybka cannot either grasp (and "catch" only later).

"ALWAYS one step back"?
But i was slaughtered on a different thread for suggesting that Rybka or Houdini on PC could be any where near as strong as Rybka cluster, and that a result of 3.5/4 to Houdini is a RIDICULOUS reason to say the cluster is weaker, since it is well within the margin of statistics that a much weaker engine can win several game in a row against a much stronger one, because in the end it will ballance out.

(I'm a bit over-reacting, because of an argument [with others] on another thread).

[Lavir]

Sure, they are still 4 games only, however from the evaluations and looking at the games you can clearly see what I was talking about.

It's true that with 4 games only everything can happen, but it's also true that if you look at the games it seems Houdini is like a GM playing against an FM (or a similar comparison).

4 games can mean nothing or everything; in this particular case I would without problems place a bet that they are actually a good representation of what it will happen in a match with many games between the two.

[Houdini]

Seven moves later, after move 38.Qg1.

[d]6r1/3r3p/p4k1P/1p1qp1p1/3n4/P2B1P2/1PP3R1/1K2R1Q1 b - - 0 38
Material still equal, but Houdini +2.43, Rybka +2.01. No idea why, white doesn't appear to be winning material.
Is this a 2 pawn positional advantage for the awkward position of the Black King?

[Dr.Wael Deeb]

Seven moves later, after move 38.Qg1.

[d]6r1/3r3p/p4k1P/1p1qp1p1/3n4/P2B1P2/1PP3R1/1K2R1Q1 b - - 0 38
Material still equal, but Houdini +2.43, Rybka +2.01. No idea why, white doesn't appear to be winning material.
Is this a 2 pawn positional advantage for the awkward position of the Black King?

No idea why

Are you kidding me

The black king's shelter is destroyed with the the king himself being shifted slowly to the center of the borad........

It's pretty obvious actually.......
Dr.D

[S.Taylor]

Sure, they are still 4 games only, however from the evaluations and looking at the games you can clearly see what I was talking about.

It's true that with 4 games only everything can happen, but it's also true that if you look at the games it seems Houdini is like a GM playing against an FM (or a similar comparison).

4 games can mean nothing or everything; in this particular case I would without problems place a bet that they are actually a good representation of what it will happen in a match with many games between the two.

_I_ fully agree with you. esp if ones sees it from the nature of the game, also.

[Lavir]

It can be. Sincerely I, as you, have no idea why the evaluation is so high. I would have not the minimal idea on how to continue to win easily.

[Dr.Wael Deeb]

On another note,Rybka Cluster is a huge disappointment unfortunately....

I deeply wished it could do better,but it didn't and I think it won't....
Dr.D

[Lavir]

It's pretty obvious actually.......
Dr.D

It's obvious that the king is in an awkward position, but would you be able to win easily against a computer (or even an human) this position (the fatal word here is *easily*, because +2.5 means almost a piece more)?

I would not, sincerely. While I can see that white is clearly better, a full continuation to win easily the game I cannot see.

This is +2.5, not +1.00.