S.Taylor wrote:Ovyron wrote:I recommend you take a look at the freeware engines, these are being developed and they have their own playing style.
maxchgr wrote:maybe the greatest achievement would be an engine in the top of the list that attacks
It's a flaw on the ELO formula itself. These engines at the top are the best at drawing, so an attacking engine that is not successful at beating them only risk losing and the formula makes it go lower on the list.
Thank you!Ovyron,
This is something I've been asking about for years, and no one ever took any notice about this subject, and made me feel like I must be out of my mind or something!
I keep on asking if it might not be that an engine which has greater insight might get more worthy wins at a higher standard even than Rybka, even if it doesn't make it to number one, because it is also more exposed to flaws.
Such an engine might play greater chess than Rybka, and have better knowledge and understanding and tactics, but just be outsmarted on some technicalities.
If I knew which engines are like this, then, that is plenty reason to prefer it over Rybka, or atleast to want it as well, Rybka not being enough to be relied on to find the best move in some positions.
If no one else, I'd think that atleast correspondence players would have appreacited my question which keeps coming up.
(I DID once have an argument about this with Vas himself, however [never did anyone else ever relate to this subject], but he didn't seem to agree with me. He seemed to be sure that anything which looks interesting or more "human" or sacrificial, if it gets lower results, then it must be weaker in its actual moves.
Actually, he seemed repulsed by the idea that brilliant sacrificial play is worth anything inthe world, if it sometimes can be refuted, by the top machines at defense. But according to him, Tal must have been dirt! Most of Tals play would probably have lost to Rybka, certainly more than other GM's of the Tal's elo. So Vas would say Tal was simply a low rated player, like perhaps 2200 elo. Well, why not? Tal's results might have shown that even in his best days!, however, against other humans, he would now be about 2700. OK, so make it 2400 against Rybka. )
So I suppose I give up then.