Computer Chess Club Archives


Search

Terms

Messages

Subject: Win at Chess suite

Author: Robert Hyatt

Date: 12:04:08 01/12/98


I have just finished cleaning everything up, relative to the null-move
mate extension and hash table refinement discussed here.  As a sanity
check, I decided to once again run win at chess at one minute per move
to see how Crafty would do on my Pentium Pro 200.  The best for an
"official" version has been 297, with it typically staying around 295
or so.

here is the "crosstable" for version 14.4:

       0  20  40  60  80 100 120 140 160 180 200 220 240 260 280
   +------------------------------------------------------------
 1 |   0   2   0   0   0   2   0  44   0   0   0   0   8   0   0
 2 |   0   0   0   0   0   0   0   0   0   1   0   9   0  15   0
 3 |   0   0   0   0   0   0   0   0  --   0   0  54   0   0  10
 4 |   0   0   0   0   0   0   0   0   0   0   0   0   0   4   0
 5 |   0   0   0   0   0   0   0   2   0   0   0   0   0  49   0
 6 |   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0
 7 |   0   0   0   0   0   0   0   0   0   0   0   0   3   0   0
 8 |   0   0   0   0   0   0   0   0   0   0   0   1   0   0   1
 9 |   0   0   2   0   0   0   0   0   0   0   0  24   0  57   0
10 |   0   3   0   0   0   0   1   0   0   1   0  --   0   0   0
11 |   0   0   0  14   8   0   1   0   0   0   0   0   0   0   8
12 |   0   0   0   0  51   0   0   0   0   0   0   0   5   0   0
13 |   0   0   0   0   0   0   0   0   0   0   1   0   0   0  33
14 |   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0
15 |   0   0  14   0   0   0   0   9   0   0   0  19   0   0   0
16 |   0   0   0   0   0   2   0   0   0  12   0   0  12   0   0
17 |   0   0   0   0   0   0   0   0   0   0   0   8   0   1  16
18 |   0   0   0   0   0   0   0   0   3   0   0   1   0   0   0
19 |   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0
20 |   0   0   0   0   0   4   0   0  22  22   0   0   0   0   0
incorrect: 163 230
sum of times(squared)=46689

The bottom line is Crafty v14.4 misses two (163 and 230) in one minute
per move.  163 is solved in 1 minute on an alpha/500, but takes just
over a couple of minutes on the P6/200.  230 I don't know about, but I
think it will take longer than that, as on the alpha Crafty still did
not get it last time I tried.

Other interesting trivia.  With a limit of 30 seconds, it misses an
additional 6 on the P6/200, although on the alpha/500 it still gets
these right but does then mist 163.

So the simple results are:

    time/move          15        30          60
    correct (P6)      287       292         298
    correct (alpha)   294       298         299

Which represents the best results I have gotten so far, discounting
Cray Blitz of course.  Cray blitz solves 297 of 300 in < 1 second
(as I have mentioned, that program times to the nearest second and
reports times of 0 for 297 of the positions).  It takes 1 second (or
a little more, but not 2 seconds) to solve 230, 92 and 222.  But this
test was run on a *big* Cray and averaged about 5M nodes per second for
the entire test suite, which burned 32 processors pretty well.  :)

For the two I miss now, 163 is the easier to solve, and simply needs a
bit more speed.  Null-move and razoring adversely affect some of these
positions, as I have solved some of them quicker in the past.

Sorry, but I don't have any times for a PII/300 yet, although we do have
one.  My P5/233mmx notebook misses two more, as it is some 15% slower
than my P6/200.

Any other results to compare with?  Be interesting to see who solves
what quickly and who has trouble with what.  I'm going to experiment
with the chess middlegame suite Bruce mentioned.  The only thing I don't
like is the length of time it takes to run a suite that large...



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.