Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Simple quad-opteron test

Author: Tord Romstad

Date: 07:46:29 12/04/03

Go up one level in this thread


On December 03, 2003 at 17:25:29, Dann Corbit wrote:

>On December 03, 2003 at 17:04:09, Tord Romstad wrote:
>
>>On December 03, 2003 at 16:07:09, Dann Corbit wrote:
>>
>>>You can save a lot of time running WAC by just running these.  The others are
>>>all trivial for all but the weakest engines:
>>
>>(List of positions snipped)
>>
>>I think the whole suite is useful, mainly in order to make sure you haven't
>>broken something when you have made a non-trivial change somewhere in the
>>search.
>>
>>Is there any system in the order in which you list the positions?
>
>Look at the position strings lexically.

Aha, of course!  I should have noticed.   :-)

>>It
>>all looks completely random to me.  For what its worth, here they
>>are in increasing order of difficulty (measured by number of positions to
>>solution) for my engine:
>
>It is a list of often missed positions.  Some of them will be very easy for many
>engines.

Of course.

>>Number  Plies  Time    Nodes
>>141:     7     0.29    52911
>
>This is impressive.  WAC.141 is one of the harder ones for many engines.

I have checks in the qsearch and extend on mate threats in the main
search, which of course helps a lot in WAC 141.

>>100:     ?        ?        ?
>>230:     ?        ?        ?
>
>Very few engines get the last 2.

OK, it seems I shouldn't worry too much about these two, and try to
be satisfied that I solve most of the others rather quickly.

>In WAC.230, there is some question as to whether it is even a solution.
>However, it is the only move with any winning chances.

I didn't know.  Thanks for the information!

Tord



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.