The secret of Houdini

[Strelkaman]

Nice to meet you Robert.I agree with your complete analysis to your point of view.I opened task manager,i saw that Houdini was running in normal priority and then i did this.I taskmanaged Critter 1.4a and set it manually in high priority.Critter was faster than in normal priority and beat Houdini by 2 points.Then i did another test.I taskmanaged Houdini and set it in a high priority.Houdini had never changed his behaviour.Same speed as before.So why Houdini doesn't respond manually to a high priority session and doesn't go faster than before?

As your other topics in the forum demonstrate, you seem to jump to conclusions based on too few games (less than 20). Even with 100 games the confidence interval is about +-50 Elo. To detect reliably a 20 Elo change you need to play at least 1,000 games.
Also, if you observe any difference by changing the process priorities of engines, it probably indicates that you're running too many other tasks in Windows while performing the engine tests.

Yes.I answered to someone else that its too early for conclusions.But i think Strelka 55.1 is faster than Strelka 55.I understand your 1000 games rule.Anyway iam looking forward to test Houdini 3.0 someday:)Thanks

[Sedat Canbaz]

The bellow standings are based

Hi Sedat,

Please, spell it below, not bellow as you always do!...

Thank you Ernest,but unfortunately my English is not so good

And be sure that next time i will always write below

Btw,it seems that you are regular reader of my postings and i hope this time i have no any grammar mistakes

Best,
Sedat

[ernest]

Btw,it seems that you are regular reader of my postings

Yes I am
and usually don't care about syntax or grammar mistakes, I just didn't like that bellow...

[Sedat Canbaz]

Btw,it seems that you are regular reader of my postings

Yes I am
and usually don't care about syntax or grammar mistakes, I just didn't like that bellow...

Hello dear Ernest,

Yes...you are right

To be honest, i studied English language without Teacher or i did not take any English course

Actually i learned my English only by using dictionaries and later i improved my English by conversation/practice with tourists from different countries

Really i apologize....

Btw,the below data is meaning for Bellow:

(v. t.) To emit with a loud voice; to shout; -- used with out.
(v.) To make a hollow, loud noise, as an enraged bull.
(v.) To roar; as the sea in a tempest, or as the wind when violent; to make a loud, hollow, continued sound.
(n.) A loud resounding outcry or noise, as of an enraged bull; a roar.
(v.) To bowl; to vociferate; to clamor.

Best,
Sedat

[geots]

The secret of Houdini chess engine is that it uses all the physical memory of your computer.It's not a chess engine as the other ones but a speedhack engine.Simply when you open task manager you see houdini is running.But you don't see that houdini is running in high priority.This is the internal Houdini hack.Make an experiment.Try to arrange an engine match between Critter 1.4a and Houdini 2.0C on Arena interface for example.Before starting the match,go to your task manager and switch Critter to high Priority.Then start the match.Critter is superior in speed too and manages to beat Houdini with a narrow margin

Welcome to the forum, and I hope you enjoy your time here. As I told the last new member, it is always wise to pick your battles very carefully. People generally like to pick ones they can win. Maybe human nature. But that is why you will see no one here eager to engage Robert in one. They don't like to lose. Again, I hope you enjoy the forum. It is much better now with our current moderation team.


All the Best,

gts

[Tennison]

To avoid this problem when testing, I think it's simple : just test with same depth and not with a time control.

I just try a fast test on 100 games at 8 Ply depth for each engine ...

Better is to test with at least depth 20 but too much time necessary for this (I can only do these long test during the night).

Code: Select all

Gui = Fritz 11
Book = HS 7 moves + (ctg)
Hash = 64mb
                        
1   Critter 1.4 64-bit    +40/-18/=42 61.00%   61.0/100  (+78 elo)
2   Houdini 2.0b Pro x64  +18/-40/=42 39.00%   39.0/100  (-78 elo)

And yes it seems very strange ... but need more games and more depth to conclude anything ... If someone can test it ...

[hgm]

What else do you hope to learn from such tests other than which engine lies most about its depth?

Have you tried TSCP against Houdini at 8 ply?

[Strelkaman]

To avoid this problem when testing, I think it's simple : just test with same depth and not with a time control.

I just try a fast test on 100 games at 8 Ply depth for each engine ...

Better is to test with at least depth 20 but too much time necessary for this (I can only do these long test during the night).

Code: Select all

Gui = Fritz 11
Book = HS 7 moves + (ctg)
Hash = 64mb
                        
1   Critter 1.4 64-bit    +40/-18/=42 61.00%   61.0/100  (+78 elo)
2   Houdini 2.0b Pro x64  +18/-40/=42 39.00%   39.0/100  (-78 elo)

And yes it seems very strange ... but need more games and more depth to conclude anything ... If someone can test it ...

INTERESTING approach by Tennison.Let me think it for a while.What Tennison did is that he locked the plies for both engines.Houdini can't maximize his abilities because he can't use the CPU memory through the advance of plies.My theory is that Houdini absorbs CPU processor memory abilities faster than any other engine.Thats why the processor is overheating faster when Houdini is running.Well Done Tennison.Iam looking forward to see other tests from you


Nice to meet you George Speight.

[Graham Banks]

Thats why the processor is overheating faster when Houdini is running.

You should try Zappa Mexico II. Never noticed your overheating problem with Houdini, but certainly did with Zappa!

[Houdini]

To avoid this problem when testing, I think it's simple : just test with same depth and not with a time control.

I just try a fast test on 100 games at 8 Ply depth for each engine ...

Testing different engines using fixed depth is meaningless. Depth 8 for Houdini is not the same as depth 8 for Stockfish is not the same as depth 8 for Rybka.
The only meaningful test condition is identical time control.