Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Nolot test fun!

Author: Bruce Moreland

Date: 15:44:37 01/13/00

Go up one level in this thread


On January 13, 2000 at 01:11:19, Jeremiah Penery wrote:

>On January 12, 2000 at 04:27:46, Shep wrote:
>
>>On January 11, 2000 at 16:36:23, Ernst A. Heinz wrote:
>>
>>>Hi Bruce,
>>>
>>>>Let's have another thread about the Nolot positions.
>>>
>>>Yeah -- that is always serious fun!
>>>
>>>>Here are my results as of now:
>>>
>>>I put the results of "DarkThought WCCC'99" on a 500MHz Compaq
>>>Alpha-21264 XP1000 right behind yours in the table below.
>>>
>>>"DarkThought" searched each position for 15 min. The "(d = ...)"
>>>information specifies the iteration depth at which the program
>>>locked onto the solution move. The "[...]" data lists the move
>>>"DarkThought" prefers for the positions it does not solve.
>>>
>>>>   DT      Ferret  DarkThought WCCC'99
>>>>           4x450   500MHz Alpha-21264
>>>>   ------  ------  -------------------
>>>>1  6hr              12:25 min (d = 12)
>>>>2  2min   31sec       /       [Bf6]
>>>>3                     /       [Be2]
>>>>4  2.5hr              /       [Be2]
>>>>5  2hr                /       [h4]
>>>>6                     /       [Bd7]
>>>>7  6hr                /       [Rb7]
>>>>8                     /       [Re1]
>>>>9  9min               /       [Rxc5]
>>>>10 2min   11sec       /       [Bb3]
>>>>11 5min   48sec     04:32 min (d = 15)
>>
>>Most micros find #1 somewhere between 3 and 20 minutes (P6-233), see
>>http://sccs.8m.com/nolot.html.
>>The only program I have that was able to find #2 was Chessmaster 5555 (again
>>P6-233) - I left it running unattended, and when I came back after about 6
>>hours, it had found the move with a +3 eval.
>
>For #1, I don't think the key is finding Nxh6.  The key is to find that it is
>winning, namely by finding 2. Nf5!  If they play Nxh6 with a -1 eval or so (as
>all the programs on your page were doing), and follow up with inferior moves,
>they are still going to lose.  The test is really about finding _winning_ moves,
>and IMO, you must find that the move is winning to have really solved any of
>these positions.

PV 02:34:54.796 14    -32 [right] Nxh6 c3 Nf5 cxb2 Qg4 Qd7 Kh2 g6 Nh4 Qe8 Nc5
dxc5 Rxe6 Qf7 Nxg6+ Kg8 Nxf8+ Kxf8 Bf5

Mine finds Nxh6 and Nf5, but it doesn't see that it is winning.

bruce



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.