Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: OK, we can make a test ...

Author: Frank Quisinsky

Date: 14:12:00 07/31/01

Go up one level in this thread


On July 30, 2001 at 21:26:59, Robert Hyatt wrote:

>On July 30, 2001 at 10:44:22, Frank Quisinsky wrote:
>
>>On July 30, 2001 at 09:31:28, Robert Hyatt wrote:
>>
>>>On July 29, 2001 at 10:50:47, Frank Quisinsky wrote:
>>>
>>>>
>>>>Hi Bob,
>>>>
>>>>the same discuss then for 2 years here :-)
>>>>
>>>>I, the ex non ponderer, will say the following:
>>>>
>>>>01.
>>>>You have right if you say that with ponder = off engines have time manangment
>>>>problems, not all and not in all games but it's an important point. I know that
>>>>this is for statistics not sooo important (ELO statistic).
>>>>
>>>>02.
>>>>Match without ponder on single system with Athlon 1Ghz:
>>>>Engine A = 1Ghz
>>>>Engine B = 1Ghz
>>>>
>>>>Match with ponder on a single system with Athlon 1Ghz:
>>>>Engine A = ~ 497-500Mhz
>>>>Engine B = ~ 497-500Mhz
>>>>
>>>>Now we have 25-40% ponder hits if the engines play on the same level (after my
>>>>experiments).
>>>>
>>>>Result:
>>>>500Mhz + 25-40% ponder hits = ~700MHz.
>>>>
>>>>With ponder = on, the matches are running with 700Mhz on 1Ghz Athlon!
>>>>With ponder = off, the matches are running with 1Ghz on 1Ghz Athlon!
>>>>
>>>>I believe better is to play with ponder = off on single system. OK, the time
>>>>managment is a good point so say please with ponder but 300 MHz is a better
>>>>point to say please without ponder.
>>>>
>>>>03.
>>>>4-piece and 5-piece tablebases and engine-engine matches on a Dual system ...
>>>>
>>>>This is not very clear.
>>>>We have engines witch used tablebases very aggressive or not very aggressive.
>>>>
>>>>Example:
>>>>Gromit - Patzer with 5-pieces ...
>>>>
>>>>Gromit played move 50 in 2:45 with ponder = on and 5-pieces.
>>>>Only 20-30% processor time in this game if Gromit used 5-pieces.
>>>>Patzer played move 50 in 1.25 with ponder = on and 5-pieces.
>>>>
>>>>Now the same positions with 4-pieces ...
>>>>
>>>>Gromit play move 50 in 1.58! with ponder = on and *4-pieces*.
>>>>Patzer played move 50 in 1.13 with ponder = on and *4-pieces*.
>>>>
>>>>Without 4-piece tablesbases are the differents not very important.
>>>>
>>>>In my opinion it is better to play eng-eng matches with ponder on a dual system
>>>>with 4-piece tablebases.
>>>>
>>>>Best
>>>>Frank
>>>
>>>
>>>Your math is fine.
>>
>>Hi Bob,
>>
>>good, for two years you say my math is bad, but in this case you have right (I
>>mean for two years) :-))
>>
>>But your reasoning is wrong.  Which would you rather do:
>>
>>>(1) play a match between two programs, using their strongest settings, and using
>>>two 700mhz processors;
>>
>>>(2) play a match between two programs, using weaker settings on at least one, if
>>>not both, using two 1ghz processors?
>>>
>>>
>>>I vote for (1) because of using the strongest settings.  You are voting for (2)
>>>to avoid wasting compute cycles.  I think (1) is more important...
>>
>>Yes, this is a good idea and a nice experiment.
>>I have an Dual Pentium III 733 MHz and an Dual Pentium III 1GHz.
>>
>>But it is not 100%ig clear which test exactly!
>>
>>What do you think about the following match idea:
>>Only with one CPUs to point 1-4.
>>
>>01.
>>Crafty 18.10 on Dual Pentium III 733Mhz with ponder against Yace 0.99.50 on Dual
>>Pentium III 1.05GHz (the machine is running with FSB = 140) without ponder, 40
>>moves in 40 minutes, 50 games.
>>
>>02.
>>Yace 0.99.50 on Dual Pentium III 733Mhz with ponder against Crafty 18.10 on Dual
>>Pentium III 1.05Ghz without ponder, 40 moves in 40 minutes, 50 games.
>>
>>03.
>>The same match but Crafty 18.10 without ponder (1.05Ghz) and Yace 0.99.50
>>(1.05Ghz) without ponder, 50 games.
>>
>>04.
>>After this matches, a new match Crafty 18.10 (1.05Ghz) with ponder - Yace
>>0.99.50 (1.05 Ghz) with ponder, 50 games.
>>
>>In 2-3 months, after my CCE tourney I can make a test.
>>If you have a better idea please write.
>>
>>I think 128 MB for hashtables, and 4pieces tablebases is good enough for the
>>experiment. Tablebases with 4Mb cache. No lean options in configuration files.
>>
>>Best
>>Frank
>
>
>I don't think the experiment is valid...  Because you don't know how much
>crafty and yace are affected by ponder=off.  It is possible that both do a bit
>worse on time management.  So that the matches match pretty well in score.  But
>does that mean ponder=off is then ok?  Only for those two programs.  You have to
>repeat this for _every_ pair of engines you want to test like that...

Hi Bob,

OK, I think I understand!

If I have more time (in 2-3 months, must play the latest rounds in my CCE
tourney) I have now new material :-))

I write a mail in 2-3 months, befor I started.
Hope you have time to check later all important points of this experiment,
better "SMP" eyes system :-).

Best and thanks
Frank



This page took 0.02 seconds to execute

Last modified: Thu, 07 Jul 11 08:48:38 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.