Author: Thorsten Czub
Date: 01:33:57 01/29/04
Go up one level in this thread
On January 29, 2004 at 03:22:31, Thomas Mayer wrote: >No, he isn't. yes - he is. >this example is idiotic -> it is a fair test. all cars crash from the same high, 10.000 feet. The conditions are the same. it's only a fair test. We will drive them afterwards and see which car drives better. >can we at least agree that the engines play on HIS >computer also chess ? let's say they try to. but if you take a chess program that was developed on a 3 Ghz PC and put it on a 450 P3 and let it blitz, you create an environment that is very different from the purpose the program was designed for. the cars were designed to DRIVE. not how they fly. so by measuring how they fly you do not measure very precise how good they drive. and this is the problem IMO in his case. > He does not test something totally different. He just >tests with another environment. > In your case you do not test how the cars drive >you test how good they can fly... which is senseless right. good that we agree on this. ok, then another example: you test all kind of cars how good they drive when driving backwards with high speed. I am sure this will help us to order the cars ... >because they are not >created to fly. but the todays chess programs are created to play blitz in 30" games ? or blitz in 5 on a p3/450 ?? aha. i am looking forward to see the next ICGA computerchess-championship done in this time control. I am sure we will get interesting results :-)) maybe this helps fritz to win :-)) >But of course it makes sense to test cars how well they are >doing at a speed of say 30 miles per hour or 150 miles per hour... Everybody is >doing that. You may understand the difference ? exactly. you test cars WITHIN A RANGE of sensible speed. you do not test them e.g. at a speed of 3 miles per hour, or at a speed of 450 miles per hour. Now you get it. My critic is that he tests 30" blitz level ! and covers this behind 60 games ! IMO he could have played 1000 games and the results would not help us more. because the level was IMO out of the range that is senseful to test. Now we can agree. I am happy that you see this now. >Besides: you might not be aware of the facts, but still MANY persons have PCs >like this PIII/450 - i am having 2 x 400 mhz AMD too. you don't tell me something new. >I am quite sure that more versions of Fritz, Shredder or >whatever run on such hardware then on fast comps with a speed above 2.0 GHz. >That is something we should not forget while testing... no of course not. we should concentrate more on bean counting 30" time controls. As i said: i am looking forward that the next ICGA championship is been done on 30" blitz. We could play the whole championship in ONE day ! amazing. :-)) >it is only destructive -> you declare his tests as nonsense... but they make >sense - as I said, the engines still play chess. more or less. >It would be even enough if they >only make sense to him... besides -> you remember that Kasparov lost once >against Genius on P90 ? kasparov was paid. it was a deal. karpov made a draw against mephisto III. it was a deal. He even commented MK1 or played games with weaker machines. this was a deal. they got money for doing so. >With 25' minutes per side ? This is nearly exactly the >compareable timecontrol to the one he used on his faster hardware... Did Genius >play senseless in those games ? Was it not able to find decent moves ?Was >Kasparov so weak ??? kasparov is a person who likes money. thats why he left fide. to make more money. you give him money and he is doing what you want when the amount of money is high enough.
This page took 0.01 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.