Author: Uri Blass
Date: 03:24:50 06/20/03
Go up one level in this thread
On June 20, 2003 at 05:50:11, Rolf Tueschen wrote: >On June 20, 2003 at 05:38:03, Uri Blass wrote: > >>On June 20, 2003 at 05:20:47, Rolf Tueschen wrote: >> >>>Computer Chess came out of the science "computer sciences". Later scientists and >>>becoming scientists came together and made a little tournament with their >>>machines. They found rules that were enough for them - because they were >>>basically scientists, so never they would have cheated each other. They had a so >>>called gentleman's agreement about possible cheats. >>> >>>Now let's stop the historical summary for a moment of thought. >>> >>>As I wrote computer chess has no inborn rules against cheating. More, it is >>>technically impossible to prevent cheatings. As long as scientists are >>>participating that is no big problem, but what happens if people participate who >>>simply have no idea what science is? We get a real problem. All kind of private >>>routines are presented with their private results although that can't be >>>accepted as scientific procedures. The answer is, privately we can do what we >>>want, science is for labs. This is a gross misunderstanding. Simply because back >>>through the bathroom window these same people claim that their results have >>>validity. But that exactly implies science because without certain exact >>>procedures you can't get validity of your data. So that is leading you into a >>>deadly circle. >>> >>>Scientists get their income from scientific institutions. Look at Bob who gives >>>his Crafty for free but who gets enough money as Professor. Now we have certain >>>people without such an income who therefore use business technology. Now where >>>is the scientific control here? As you know software in general is a fine >>>medium. Errors are called bugs and sold as if - they had no bugs, but if they >>>have, the users give precious feedback for the business companies. In short >>>there is no scientific control whatsoever. Brilliant for the business companies. >>>They are mainly amateurs (and Christians in the majority) who do a charity job >>>for the million users. The products (programs) are tested by - again - amateur >>>testers. So all without validity. All without a way to complain if something >>>goes wrong. >>> >>>Can you follow me what I mean if I say that non-scientists, amateurs and charity >>>people sell something that we should NEVER expect scientific reliability? Not to >>>speak of validity. Excuse the many scientific terms. >>> >>>Can you also follow me that if such amateurs want to make money, NB that >>>Kasparov or Amir Ban got thousands of dollars for their show event meant as PR >>>action for the ChessBase program Junior, that then they must create a bit of hot >>>air, they must "make a little cheat" about the content of the box they are >>>selling? Of course they must say that Junior is GM!! Since Kasparov said it. Of >>>course they must shout, that the original engine that played KASPAROV IS IN THE >>>BOX!! If they didn't they were bad amateurs or - - well, just scientists. But >>>since they aren't all is kosher. >>> >>>Look, when I bought Fritz 8 I suffered of the same mental attack all the Junior >>>8 customers suffered from, I believed that I could finally use the new feature >>>with the 3D pieces. I did NEVER think about my old PC who simply had not the >>>modern graphics which were necessary to be able to profit from the new features! >>>The same with Junior 8. Against Kasparov the prog ran on extremely expensive >>>hardware. Obviously nobody around has such a machine. So by force nobody can use >>>the exact program that played Kasparov. But that was exactly what the PR of >>>ChessBase told us. But for real computer freaks - is that a surprise?? Is that a >>>cheat?? Of course NOT. Since we are totally out of science. >> >>The fact that nobody has the hardware does not mean that nobody can use the >>exact program that played kasparov. >> >>If the same program can run on slower hardware then it means that people can get >>the same program and expect it usually to play the same moves if they give it >>enough time. > >Astonishing that you make that mistake. The indeterminism by parallelism is >still leading to a non-comparable situation if you let time pass on you >1-processor machine. > >Rolf I used the words usually. I know that more than one processor may give sometimes slightly different results(different hash tables may also lead to different results) but if the evaluation is the same then I expect more than 90% of the moves to be reproducable with enough time. Uri
This page took 0 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.