Author: Robert Hyatt
Date: 08:38:14 08/30/01
Go up one level in this thread
On August 30, 2001 at 10:07:40, Uri Blass wrote: >On August 30, 2001 at 09:54:30, Robert Hyatt wrote: > >>On August 30, 2001 at 09:38:16, Gian-Carlo Pascutto wrote: >> >>>On August 30, 2001 at 09:36:26, Uri Blass wrote: >>> >>>>I posted the position some weeks ago and yace also could see the mate >>>>after going forward and backward but not from the root position. >>>> >>>>Uri >>> >>>I have fed your position to Sjengs matesolving mode too but it >>>didn't show a sign of converging on a solution after more than >>>an hour or so. >>> >>>-- >>>GCP >> >> >>Then perhaps this is one position that shows the power/speed of the _old_ >>hardware? > >I do not see what it has to do with the old hardware. >this position was from my correspondence game and I mentioned it >only because mate in 18 was mentioned. > >The position has nothing to do with Deep blue-deep thought games. I was talking about the mate in 18 from a deep thought game. If programs of today can't find it, but DT could, that is at least one example of how their search performs in certain types of tactical situations... If I misunderstood what you were talking about, then forget it as it would be useless since we don't have access to DT to see how it would compare to today's programs in a position like this. I had thought that we _already_ knew that DT saw the mate in 18, but that others could not find it. > >Here is the position again for people who are interested > >Abir Har aven - Uri Blass >[D]2k1r3/ppp5/5P2/8/3P1P1q/3P2Nr/PP1Q1K1n/6RR b - - 0 1 > >The solution begins with >26...Ng4+ 27.Kf1 Ne3+ 28.Ke1 and only finding this main line is a big >problem for a lot of program and they usually like 28.Qxe3. > >Finding the solution after 28.Ke1 is also an hard problem and I did it >only by a tree because programs cannot see even after >28.Ke1 a forced mate from the root position. > >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.