Author: Ed Schröder
Date: 05:15:04 01/28/02
Go up one level in this thread
On January 28, 2002 at 07:42:39, Amir Ban wrote: >On January 28, 2002 at 07:33:59, Albert Silver wrote: > >>On January 28, 2002 at 07:29:03, Uri Blass wrote: >> >>>On January 28, 2002 at 07:11:05, Albert Silver wrote: >>> >>>>On January 28, 2002 at 06:58:25, Uri Blass wrote: >>>> >>>>>On January 28, 2002 at 06:42:25, Albert Silver wrote: >>>>> >>>>>>On January 28, 2002 at 06:33:26, Amir Ban wrote: >>>>>> >>>>>>>On January 28, 2002 at 06:12:53, Ed Schröder wrote: >>>>>>> >>>>>>>>>Will TACTIC's eventually REFUTE! Positional play? >>>>>>>> >>>>>>>>In the end yes. >>>>>>>> >>>>>>>>It is my (new) opinion that the nature of chess is just search. >>>>>>>> >>>>>>>>Elo progress of (professional) chess programs... >>>>>>>> >>>>>>>>1990 - elo 2000 (average depth 6-8) (TC 40/2h) >>>>>>>>1995 - elo 2300 (average depth 8-10) >>>>>>>>2000 - elo 2500 (average depth 11-13) >>>>>>>>2002 - elo 2600 (average depth 12-14) >>>>>>>> >>>>>>> >>>>>>>This begs the question, because the programs are newer and play positionally >>>>>>>different. Will a 1990/1995 program perform 2600+ on today's hardware ? >>>>>>>Doubtful. >>>>>>> >>>>>>> >>>>>>>>No way to stop it. >>>>>>>> >>>>>>>>No suprise Kasparov lost against Deep Blue. >>>>>>> >>>>>>>It was a surprise because he is clearly better. >>>>>>> >>>>>>>> >>>>>>>>The sad future: it will be in the headlines when a grandmaster occasionally will >>>>>>>>win from a computer. >>>>>>>> >>>>>>> >>>>>>>If this will happen due to positionally outplaying will you also consider it sad >>>>>>>? >>>>>>> >>>>>>>Amir >>>>>> >>>>>>I don't think he meant it would be sad they won, but that this would be >>>>>>considered news. In any case, I'm afraid I agree with him on the nature of >>>>>>chess. I think that positional play is just extremely deep and refined tactical >>>>>>play. Since we approach them differently, we regard them as different, but that >>>>>>is still how I regard them. Notice how already some elements of knowledge that >>>>>>were necessary in older programs are removed as the search makes up for it. >>>>> >>>>>I do not believe in it. >>>>> >>>>>If it is the case palm tiger should have knowledge that the default tiger does >>>>>not have but it is not the case because I remember from christophe's posts that >>>>>palm tiger is the same engine that he is using for tournaments except more hash >>>>>tables book and better hardware. >>>>> >>>>>Palm tiger is better than the old programs on similiar hardware and it means >>>>>that exactly the same knowledge that is good for today programs can be also good >>>>>for old programs. >>>>> >>>>>Uri >>>> >>>>I was speaking about the reverse: knowledge that was necessary or useful in >>>>older programs with much more limited hardware but that is not necessary or >>>>useful today. >>>> >> >>>I understood and my point is that knowledge that was used in palm tiger is also >>>necessary for tiger on good hardware and palm tiger is the best for it's >>>hardware. >>> >>>Uri >> >>There used to be knowledge that was deliberately inserted to avoid things such >>as giving a piece to play a fork and capture the rook on a1/a8. This was useful >>because they always played this (the Mephisto MMIV did IIRC) and lost the second >>piece (the knight) a bit after. Today this horizon effect is compensated by the >>depth of the search. This is just an example, and I'm sure the programmers could >>say more about this. > >A white knight on a8/h8 is weak and often in trouble, and there's nothing wrong >with factoring that in the evaluation function. The fact that search will often >find this on its own is immaterial. There is nothing wrong having this kind of knowledge in your eval. The BIG question is: IS IT NEEDED as search can do that job for you too? If a deep search depth guarantees you to deal with the Na8/Nh8 cases why do you want to do things twice and spoil valuable processor time? Ed >Look at CCT4 ZarkovX vs. DJ for a nice demonstration. > >Amir
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.