Houdini 3

[marijan]

Hi Robert,

Don't forget to dedicate some resources for testing at long time controls too as some people here are implying Houdini 1.5 is stronger than version 2 at the regular OTB time control (40 moves in 120 minutes).

Best regards

Mel, some of the most vocal of the "some people here" have never actually tested Houdini 2.....

Robert

Well said Robert!

Here is results from my 60 min per engine test...

i3 370 M
2 cores
512 hash
reversible starting pos 8 moves

[lech]

I really like to watch games of Houdini 1.5a. Strategy, through good evoluation of positions, is high. It seems logical that the other engines on the longer TC (depths of search) should partially reduce this loss.

[Rebel]

Congratulations for the +25 ELO upon last release: I am struggling to even get +5 ELO (although I have access to Critter sources that Richard was very kind to send me) so I understand very well what it takes to increase an already top engine.

LMR seems to be exploited to its limit and progress seems to stagnate. Basically we are waiting for the next breakthrough such as null-move and LMR were.

[mcostalba]

LMR seems to be exploited to its limit and progress seems to stagnate. Basically we are waiting for the next breakthrough such as null-move and LMR were.

The hundreds of ELO of difference between Galurung and Stockfish are the sum of a lot of small and (very rarely) medium gains spread across the whole engine: I don't believe in silver bullets

[lkaufman]

Hi Robert,

Don't forget to dedicate some resources for testing at long time controls too as some people here are implying Houdini 1.5 is stronger than version 2 at the regular OTB time control (40 moves in 120 minutes).

Best regards

Mel, some of the most vocal of the "some people here" have never actually tested Houdini 2 . If you have any doubts about Houdini 2 at long TC, check the neighboring thread, or the very interesting matches of various engines against Deep Junior 13. Houdini 2 scores about 10% better than its closest competitors.

As for Houdini 3, it will do quite well at long TC - some search algorithm improvements will have an increasing effect with higher search depths.

Robert

The first test you cite does not have either Komodo or Houdini 1.5, so it doesn't tell us anything about whether Houdini 2 scales as well as either of those two engines.

Of course, Komodo is the best scaling program out there. It is absolutely unthinkable that it scales as well as anything else, just has some problems with fast TC.

Comparing to Critter and Ivanhoe just confirms my contention that neither of these scale any better than Houdini as they all have basically the Ippo search, with various tweaks.

Komodo has basically Ippo search too. (alpha-beta + null-move + IID + aggressive lmr)
Edit: ...with various tweaks of course

Leaving Critter aside (so you won't take this as directed at you), it is clear that Komodo is weaker than Ivanhoe (or any acknowledged Ippo derivative) at bullet chess or faster, but stronger at blitz and much stronger at slow chess. If we compare Komodo to the very different (from Ivanhoe) program Stockfish, we don't see this pattern, Komodo is just slightly stronger at all levels. I still don't understand the reason for this scaling behavior vs. the Ippo programs, but I would very much like to get at the reason for it. Since Komodo is not open-source, I don't expect anyone to be able to solve this puzzle, but since Stockfish is open-source as is Ivanhoe perhaps someone can comment on why SF is weaker than Ivanhoe at bullet chess but stronger at normal time controls? I think it is something to do with the search, though it could also be caused by eval. Basically I would like to know if Komodo (and Stockfish) are weak at bullet chess because we are doing something wrong, or whether we are stronger at slower chess because Ippo/Ivanhoe is doing something wrong (or both). Comments anyone?

[michiguel]

For your information, the Houdini 3 release is planned for September 2012.
See the Houdini Facebook Page http://www.facebook.com/pages/Houdini-C ... 0926948947 to find the current progress.

Cheers,
Robert

Congratulations for the +25 ELO upon last release: I am struggling to even get +5 ELO (although I have access to Critter sources that Richard was very kind to send me) so I understand very well what it takes to increase an already top engine.

On a funny side note I see that Larry Kaufmann was able to make Richard to get angry, it is not an easy result, I think he can be already satisfied with this and just leave

We need more GMs in this forum, not less.

Miguel

[MM]

Hi Robert,

Don't forget to dedicate some resources for testing at long time controls too as some people here are implying Houdini 1.5 is stronger than version 2 at the regular OTB time control (40 moves in 120 minutes).

Best regards

Mel, some of the most vocal of the "some people here" have never actually tested Houdini 2 . If you have any doubts about Houdini 2 at long TC, check the neighboring thread, or the very interesting matches of various engines against Deep Junior 13. Houdini 2 scores about 10% better than its closest competitors.

As for Houdini 3, it will do quite well at long TC - some search algorithm improvements will have an increasing effect with higher search depths.

Robert

The first test you cite does not have either Komodo or Houdini 1.5, so it doesn't tell us anything about whether Houdini 2 scales as well as either of those two engines.

Of course, Komodo is the best scaling program out there. It is absolutely unthinkable that it scales as well as anything else, just has some problems with fast TC.

Comparing to Critter and Ivanhoe just confirms my contention that neither of these scale any better than Houdini as they all have basically the Ippo search, with various tweaks.

Komodo has basically Ippo search too. (alpha-beta + null-move + IID + aggressive lmr)
Edit: ...with various tweaks of course

Leaving Critter aside (so you won't take this as directed at you), it is clear that Komodo is weaker than Ivanhoe (or any acknowledged Ippo derivative) at bullet chess or faster, but stronger at blitz and much stronger at slow chess. If we compare Komodo to the very different (from Ivanhoe) program Stockfish, we don't see this pattern, Komodo is just slightly stronger at all levels. I still don't understand the reason for this scaling behavior vs. the Ippo programs, but I would very much like to get at the reason for it. Since Komodo is not open-source, I don't expect anyone to be able to solve this puzzle, but since Stockfish is open-source as is Ivanhoe perhaps someone can comment on why SF is weaker than Ivanhoe at bullet chess but stronger at normal time controls? I think it is something to do with the search, though it could also be caused by eval. Basically I would like to know if Komodo (and Stockfish) are weak at bullet chess because we are doing something wrong, or whether we are stronger at slower chess because Ippo/Ivanhoe is doing something wrong (or both). Comments anyone?

Ippo-Robbo-Ivan derivates are extremely tactical, so easily their strenght drops at longer time control. On the other hand Komodo plays in a positional style and doesn't excel in tactics so it clearly scales better at longer time controls.

It's just like human playing.

Someone needs more time to find the right move.

Best Regards

[lkaufman]

Hi Robert,

Don't forget to dedicate some resources for testing at long time controls too as some people here are implying Houdini 1.5 is stronger than version 2 at the regular OTB time control (40 moves in 120 minutes).

Best regards

Mel, some of the most vocal of the "some people here" have never actually tested Houdini 2 . If you have any doubts about Houdini 2 at long TC, check the neighboring thread, or the very interesting matches of various engines against Deep Junior 13. Houdini 2 scores about 10% better than its closest competitors.

As for Houdini 3, it will do quite well at long TC - some search algorithm improvements will have an increasing effect with higher search depths.

Robert

The first test you cite does not have either Komodo or Houdini 1.5, so it doesn't tell us anything about whether Houdini 2 scales as well as either of those two engines.

Of course, Komodo is the best scaling program out there. It is absolutely unthinkable that it scales as well as anything else, just has some problems with fast TC.

Comparing to Critter and Ivanhoe just confirms my contention that neither of these scale any better than Houdini as they all have basically the Ippo search, with various tweaks.

Komodo has basically Ippo search too. (alpha-beta + null-move + IID + aggressive lmr)
Edit: ...with various tweaks of course

Leaving Critter aside (so you won't take this as directed at you), it is clear that Komodo is weaker than Ivanhoe (or any acknowledged Ippo derivative) at bullet chess or faster, but stronger at blitz and much stronger at slow chess. If we compare Komodo to the very different (from Ivanhoe) program Stockfish, we don't see this pattern, Komodo is just slightly stronger at all levels. I still don't understand the reason for this scaling behavior vs. the Ippo programs, but I would very much like to get at the reason for it. Since Komodo is not open-source, I don't expect anyone to be able to solve this puzzle, but since Stockfish is open-source as is Ivanhoe perhaps someone can comment on why SF is weaker than Ivanhoe at bullet chess but stronger at normal time controls? I think it is something to do with the search, though it could also be caused by eval. Basically I would like to know if Komodo (and Stockfish) are weak at bullet chess because we are doing something wrong, or whether we are stronger at slower chess because Ippo/Ivanhoe is doing something wrong (or both). Comments anyone?

Ippo-Robbo-Ivan derivates are extremely tactical, so easily their strenght drops at longer time control. On the other hand Komodo plays in a positional style and doesn't excel in tactics so it clearly scales better at longer time controls.

It's just like human playing.

Someone needs more time to find the right move.

Best Regards

I think you are right, but the question is WHY are the Ippo-group programs more tactical but weaker positionally? Is it simply that they have cruder eval that is faster but inferior, or is it that some aspect of the search makes them tactically stronger but positionally weaker? Only someone who has studied the Ippo code can answer this of course, especially if it compared to SF code, as SF has the same general behavior as Komodo in this respect.

[mclane]

Basically I would like to know if Komodo (and Stockfish) are weak at bullet chess because we are doing something wrong, or whether we are stronger at slower chess because Ippo/Ivanhoe is doing something wrong (or both). Comments anyone?

komodo and stockfish are intelligent evaluation programs.
ippo/ivanhoe or rybka etc. are stupid search programs.

the intelligent programs are weak at bullet, and better at slow chess.

[lkaufman]

Basically I would like to know if Komodo (and Stockfish) are weak at bullet chess because we are doing something wrong, or whether we are stronger at slower chess because Ippo/Ivanhoe is doing something wrong (or both). Comments anyone?

komodo and stockfish are intelligent evaluation programs.
ippo/ivanhoe or rybka etc. are stupid search programs.

the intelligent programs are weak at bullet, and better at slow chess.

I won't disagree with you, but is this based on comparing the evaluation functions in the code or just based on watching play? Is it generally agreed by people who have studied both that Stockfish has a more sophisticated/advanced evaluation function (but a slower one) than Ivanhoe/Ippo? If the explanation is this simple then there is nothing for us to do to change this behavior, but I am not sure this is the whole story.