Author: Ed Schröder
Date: 02:50:17 02/15/01
Go up one level in this thread
On February 15, 2001 at 05:12:28, Uri Blass wrote: >On February 15, 2001 at 03:56:10, Eelco de Groot wrote: > >>On February 14, 2001 at 01:39:23, Jeremiah Penery wrote: >> >>>On February 13, 2001 at 20:43:07, Joshua Lee wrote: >>> >>>>No i am just showing that this isn't tough compared to the others and i think >>>>there should be a new suite with problems not solved under 1hour but some >>>>positions are solve in 30minutes with one program and 5 with another.... >>> >>>I think it would be really neat if someone created a new set of positions to >>>perhaps replace the Nolot ones. It's known that a couple of the Nolot solutions >>>are at least slightly dubious (#9 being foremost, #3 and #6 might also be, and >>>it can be argued that the solution to #8 may not be the "best" move). Also >>>nowadays, at least #11 is too easy for computers to be in a suite of "unsolvable >>>by computer" problems. >>> >>>Maybe if I have the time I'll look for some positions where the solution is >>>really the best move, provide/find analysis for them, and make sure they're >>>really hard for computers. >>> >>>One position that comes to mind already is Shirov's 47. ...Bh3!! against >>>Topalov, Linares 1998: >>>(I hope the diagram is correct.) >>>[D]8/8/4kpp1/3p1b2/p6P/2B5/6P1/6K1 b - - >>> >>>Computers find this extremely difficult, at the least. :) >> >>Are you sure this is the correct position, Jeremiah? Q5T personality for Rebel >>Century 2.0, only 512Kb transposition tables and 5 more engines in background, >>me typing and connecting to Internet: > >Q5T was not generated to play chess and it is also going to find some wrong >combinations. >I suspect that you can find every best move by defining the right personality. >It proves nothing. > > >The point is that as far as I know no playing program can find Bh3. > >>The Nolot position No.11 it has a lot more problems with finding the best move! > >I guess that by changing parameters you can solve nolot number 11 in less than 1 >seconds. >It is possible that Q5T is not the right personality for this position. > >Uri That is maybe all true. The Q5T engine (and the default tactical engine) are perhaps playing 200 elo points less strong than the default setting it is still good enough for a 2300 elo at least. The point is the ease of finding tactical shots and hardly any wrong sacrifice made. It is something people really enjoy because most people don't care about 2300 or 2600. Of course here in CCC that is different a different story :) Ed
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.