100 long games Rybka 4 vs Houdini 1.03a

[Laskos]

Rybka 4 is weaker than latest Ippos at ANY time control, period.

Haven't seen the evidence of that, not that I really care anyway. Not everybody is interested in taking things that "fall off the back of a truck".

totally agree

Rybka4 64 bit is 5 elo better than Ivanhoe 64 bits in SWCR rating list
based on Frank Quisinsky.

http://talkchess.com/forum/viewtopic.php?t=36472

1 Rybka 4 x64 2944 18 18 1440 82% 2687 28%
2 IvanHoe T0.4 x64 2939 21 21 920 79% 2720 32%

You are strange, Uri. Do you see the error margins (sqrt(error1^2+error2^2))? Do you know what are the latest or best IvanHoes? Without talking of Houdini.

Really strange statements from you, but if you passionately dislike hidden authors of engines, then that could be explained. How do you test Movei? Do you need more than 30,000 games? Can someone prove with 30,000+ games at long TC that Rybka 4 is stronger? So, shut please up!

Kai

[Uri Blass]

Rybka 4 is weaker than latest Ippos at ANY time control, period.

Haven't seen the evidence of that, not that I really care anyway. Not everybody is interested in taking things that "fall off the back of a truck".

totally agree

Rybka4 64 bit is 5 elo better than Ivanhoe 64 bits in SWCR rating list
based on Frank Quisinsky.

http://talkchess.com/forum/viewtopic.php?t=36472

1 Rybka 4 x64 2944 18 18 1440 82% 2687 28%
2 IvanHoe T0.4 x64 2939 21 21 920 79% 2720 32%

You are strange, Uri. Do you see the error margins (sqrt(error1^2+error2^2))? Do you know what are the latest or best IvanHoes? Without talking of Houdini.

Really strange statements from you, but if you passionately dislike hidden authors of engines, then that could be explained. How do you test Movei? Do you need more than 30,000 games? Can someone prove with 30,000+ games at long TC that Rybka 4 is stronger? So, shut please up!

Kai

From the Ipon rating list:

1 Deep Rybka 4 2953 13 13 2800 79% 2727 28%
2 Houdini 1.03a 2950 13 12 2800 79% 2727 28%

I stop developing movei and it is possible that A is 20 or 30 elo better than B at fast time control when it is weaker at longer time control.

[Albert Silver]

If Rybka team of testers is not testing their engine in matches of 30,000+ games (irrespective of time controls), they are incompetent. I know that they are incompetent because Rybka 4 is weaker by quite a bit compared to lateast Ippos. Stop whining here, Rybka 4 is weaker than latest Ippos at ANY time control, period.

Quite a lot of nonsense you post. You should really stick to statistics IMHO.

1) If Rybka team of testers is not testing their engine in matches of 30,000+ games (irrespective of time controls), they are incompetent.

Well, let us start with the first beta which had severe time management issues. Apparently this would have been caught by playing 30 thousand games. We would play the match, look at the score, and BINGO! would realize this was a time management issue. Then there are a few knowledge issues. Again, play a couple of 30 thousand-game matches, and ABRACADABRA, the GUI will give a number and explain that a certain type of position is hurting its bottom line.

2) I know that they are incompetent because Rybka 4 is weaker by quite a bit compared to latest Ippos.

Aha! So the strength (or weakness) of Rybka is the fault of the *testers*!! Your mentality is ideally suited for the corporate environment where the motto is: "It is someone else's fault".

3) Rybka 4 is weaker than latest Ippos at ANY time control, period.

It is rather odd that someone who spends his time screaming "statistics" every other post blithely ignores the results of Frank or Ingo where this blanket judgment somehow doesn't seem to fit.

4) As to enjoying the games... The fact that you do not is your misfortune.

[Uri Blass]

Here is an example from the CCRL rating list when the order is dependent on the time control(it seems clear that Movei earns rating relative to Colossus when the time control is longer)

CCRL 40/40
37 Movei 00.8.438 (10 10 10) 2772 +13 −13 47.9% +11.9 38.2% 2184
40-41 Colossus 2008b 2749 +15 −15 48.7% +9.4 39.2% 1552

CEGT 40/20

459 Movei 0.08.438 P10 2673 17 17 1096 55.3% 2636 35.3%
472 Colossus 2008b 2665 16 16 1180 49.9% 2665 37.0


CCRL 40/4
40 Colossus 2008b 2749 +11 −11 42.2% +59.8 29.4% 3564
44 Movei 00.8.438 (10 10 10) 2731 +9 −9 43.0% +53.4 29.3% 4897

[Laskos]

From the Ipon rating list:

1 Deep Rybka 4 2953 13 13 2800 79% 2727 28%
2 Houdini 1.03a 2950 13 12 2800 79% 2727 28%

I stop developing movei and it is possible that A is 20 or 30 elo better than B at fast time control when it is weaker at longer time control.

Do you believe in the myth of Long Time Control? It is used mostly as a whining for performing badly at fast controls. As for Ipon, look at error margins, +/- 17 Elo points 95% confidence comparing the two engines. You are not proving anything with that. My proof was much more solid in a direct match with +/- 4 Elo points 95% confidence and LOS close to 100% at ultra short time controls. The burden is on Rybka freaks to show that their engine is stronger than the latest Ippos. My part is done.

Kai

ps Pity you abandoned Movei.

[Albert Silver]

From the Ipon rating list:

1 Deep Rybka 4 2953 13 13 2800 79% 2727 28%
2 Houdini 1.03a 2950 13 12 2800 79% 2727 28%

I stop developing movei and it is possible that A is 20 or 30 elo better than B at fast time control when it is weaker at longer time control.

Do you believe in the myth of Long Time Control? It is used mostly as a whining for performing badly at fast controls. As for Ipon, look at error margins, +/- 17 Elo points 95% confidence comparing the two engines. You are not proving anything with that. My proof was much more solid in a direct match with +/- 4 Elo points 95% confidence and LOS close to 100% at ultra short time controls. The burden is on Rybka freaks to show that their engine is stronger than the latest Ippos. My part is done.

Kai

ps Pity you abandoned Movei.

Actually, I didn't see any proof. First of all you gave zero information on the test conditions, meaning you might have tested without books, or with. The only information you gave showed you played Rybka in a handicapped 32-bit environment.

As to being stronger... In all honesty I am not at all convinced Rybka 4 is stronger than Houdini 1.03a. I think they are roughly equal.

[Laskos]

If Rybka team of testers is not testing their engine in matches of 30,000+ games (irrespective of time controls), they are incompetent. I know that they are incompetent because Rybka 4 is weaker by quite a bit compared to lateast Ippos. Stop whining here, Rybka 4 is weaker than latest Ippos at ANY time control, period.

Quite a lot of nonsense you post. You should really stick to statistics IMHO.

1) If Rybka team of testers is not testing their engine in matches of 30,000+ games (irrespective of time controls), they are incompetent.

Well, let us start with the first beta which had severe time management issues. Apparently this would have been caught by playing 30 thousand games. We would play the match, look at the score, and BINGO! would realize this was a time management issue. Then there are a few knowledge issues. Again, play a couple of 30 thousand-game matches, and ABRACADABRA, the GUI will give a number and explain that a certain type of position is hurting its bottom line.

I think that many developers of engines play 30,000+ games for each small step, as Bob does. I am sure Vas sometimes does that, otherwise it's hard to improve anything.

2) I know that they are incompetent because Rybka 4 is weaker by quite a bit compared to latest Ippos.

Aha! So the strength (or weakness) of Rybka is the fault of the *testers*!! Your mentality is ideally suited for the corporate environment where the motto is: "It is someone else's fault".

I said previously that "I got it". I got it why. If M Ansari is an important tester there watching the games, then Rybka is an endangered species.

3) Rybka 4 is weaker than latest Ippos at ANY time control, period./b]

It is rather odd that someone who spends his time screaming "statistics" every other post blithely ignores the results of Frank or Ingo where this blanket judgment somehow doesn't seem to fit.

Their statistics is weak, their choice of Ippo* engines is wrong. If you do not know that, then please inform yourself, this is all free stuff and open source.

4) As to enjoying the games... The fact that you do not is your misfortune.

I enjoy it with my Chessmaster on my phone. But judging a 3,000+ Elo engine even by Uri is ridiculous.

Kai

[Graham Banks]

From the Ipon rating list:

1 Deep Rybka 4 2953 13 13 2800 79% 2727 28%
2 Houdini 1.03a 2950 13 12 2800 79% 2727 28%

I stop developing movei and it is possible that A is 20 or 30 elo better than B at fast time control when it is weaker at longer time control.

Do you believe in the myth of Long Time Control? It is used mostly as a whining for performing badly at fast controls. As for Ipon, look at error margins, +/- 17 Elo points 95% confidence comparing the two engines. You are not proving anything with that. My proof was much more solid in a direct match with +/- 4 Elo points 95% confidence and LOS close to 100% at ultra short time controls. The burden is on Rybka freaks to show that their engine is stronger than the latest Ippos. My part is done.

Kai

ps Pity you abandoned Movei.

Alternatively we could leave it to the Ippo freaks to continue trying to convince everybody otherwise.
At least I agree with you that it's disappointing that Uri stopped developing Movei.

[tomgdrums]

From the Ipon rating list:

1 Deep Rybka 4 2953 13 13 2800 79% 2727 28%
2 Houdini 1.03a 2950 13 12 2800 79% 2727 28%

I stop developing movei and it is possible that A is 20 or 30 elo better than B at fast time control when it is weaker at longer time control.

Do you believe in the myth of Long Time Control? It is used mostly as a whining for performing badly at fast controls. As for Ipon, look at error margins, +/- 17 Elo points 95% confidence comparing the two engines. You are not proving anything with that. My proof was much more solid in a direct match with +/- 4 Elo points 95% confidence and LOS close to 100% at ultra short time controls. The burden is on Rybka freaks to show that their engine is stronger than the latest Ippos. My part is done.

Kai

ps Pity you abandoned Movei.

How come all the ippolites are always so angry? Have they been watching too much Glenn Beck?

[Graham Banks]

How come all the ippolites are always so angry? Have they been watching too much Glenn Beck?

Perhaps it's really frustrating trying to wring anything further out of somebody else's code.