Reply
 
LinkBack Thread Tools Display Modes
  #1   Report Post  
Old April 4th 04, 04:09 PM
Richard Heldmann
 
Posts: n/a
Default Crafty compilation problem?

This epd position 4k3/4p3/8/8/8/8/8/3QKB2 w - - 0 1;

is found by Crafty 19.3; it finds this mate in less than a second
using the 5-man tablebases. See below:

depth=4 +327.54 1. Qh5+ EGTB
Nodes: 55 NPS: 5500
Time: 00:00:00.01

Crafty 19.11 produces this output:
depth=0 +0.00
Nodes: 0 NPS: 0
Time: 00:00:00.01

This was done on a Pentium 400Mhz machine, using Winboard 4.2.6. Can
someone explain this to me.
Thank you.
  #2   Report Post  
Old April 4th 04, 11:00 PM
Mikko Nummelin
 
Posts: n/a
Default Crafty compilation problem?

On Sun, 4 Apr 2004, Richard Heldmann wrote:

This epd position 4k3/4p3/8/8/8/8/8/3QKB2 w - - 0 1;

is found by Crafty 19.3; it finds this mate in less than a second
using the 5-man tablebases. See below:

depth=4 +327.54 1. Qh5+ EGTB
Nodes: 55 NPS: 5500
Time: 00:00:00.01

Crafty 19.11 produces this output:
depth=0 +0.00
Nodes: 0 NPS: 0
Time: 00:00:00.01

This was done on a Pentium 400Mhz machine, using Winboard 4.2.6. Can
someone explain this to me.


I just looked at crafty-19.11 and crafty-19.12 (newest) by taking a
difference from their source codes. I noticed some changes to the egtb.cpp
-code (maybe those are bugfixes). In fact, the Nalimov tablebase
acceessing code in many different programs, including Crafty and Scid is a
bunch of very unclean and difficult-to-understand -C++-hasslings,
probably copied around and around from place to place without further
consideration. Of course the commercial programs are probably more
thoroghly checked and perhaps do use way different code for this purpose.

What we actually need is a _clean and compact_ standard description for
egtb:s, not only ad-hoc code whose function is described nowhere in any
public place in the Internet.


Mikko Nummelin
  #3   Report Post  
Old April 5th 04, 12:48 AM
jake johnson
 
Posts: n/a
Default Crafty compilation problem?

(Richard Heldmann) wrote in message . com...
This epd position 4k3/4p3/8/8/8/8/8/3QKB2 w - - 0 1;

is found by Crafty 19.3; it finds this mate in less than a second
using the 5-man tablebases. See below:

depth=4 +327.54 1. Qh5+ EGTB
Nodes: 55 NPS: 5500
Time: 00:00:00.01

Crafty 19.11 produces this output:
depth=0 +0.00
Nodes: 0 NPS: 0
Time: 00:00:00.01

This was done on a Pentium 400Mhz machine, using Winboard 4.2.6. Can
someone explain this to me.
Thank you.

Please include more information. You mention compilation problem on
the subject line but you don't expand on this. What OS's are you
using? Mate in how many? Your 19.11 output does look unusual though,
however. First is my 19.3 output from a 1200Mhz WinXP laptop:
White(1): go
time surplus 0.00 time limit 30.00 (3:30)
depth time score variation (1)
2 0.00 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
2- 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
3 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
3- 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
4 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
4- 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
5 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
5- 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
6 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
6- 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
7 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
7- 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
8 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
8- 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
9 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
9- 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
10 0.01 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
10- 7.48 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+ HT
11 11.34 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+
EGTB
time=36.11 cpu=95% mat=11 n=29311500 fh=99%
nps=811k
ext- chk=4934910 cap=9 pp=4986 1rep=118798 mate=0
predicted=0 nodes=29311500 evals=12561664
endgame tablebase- probes done=25006 successful=24980
SMP- split=0 stop=0 data=0/32 cpu=34.62 elap=36.11

mate in 7 moves.

kibitz mate in 7 moves.


White(1): Bc4
time used: 36.11
----------------------------------------------
Below is my output from 19.11 on a 533Mhz celeron running FreeBSD 4.5.
I gave it 300 seconds to reply and it was more than needed, at least
if your desired answer is Bc4. The tablebase between these two
computers may vary a bit and you may have more tablebases that I have
of course so our results aren't directly comparable.

end-game phase
clearing hash tables
time surplus 0.00 time limit 3:20 (3:20)
depth time score variation (1)
6- 0.59 Mat08 1. Bc4 e5 2. Qh5+ Kd7 3. Qxe5
EGTB
7 0.67 Mat08 1. Bc4 e5 2. Qh5+ Kd7 3. Qxe5
EGTB
7- 1.39 Mat08 1. Bc4 e5 2. Qh5+ Kd7 3. Qxe5
EGTB
8 1.64 Mat08 1. Bc4 e5 2. Qh5+ Kd7 3. Qxe5
EGTB
8- 4.65 Mat08 1. Bc4 e5 2. Qh5+ Kd7 3. Qxe5
EGTB
9 7.00 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+
EGTB
9- 15.47 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+
EGTB
10 16.61 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+
EGTB
10- 49.18 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+
EGTB
11 53.09 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+
EGTB
11- 2:21 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+
EGTB
12 2:33 Mat07 1. Bc4 e6 2. Qd6 Kf7 3. Qxe6+
EGTB
time=3:20 cpu=99% mat=11 n=55459969 fh=99% nps=277K
ext- chk=9090139 cap=0 pp=117 1rep=393998 mate=1989
predicted=0 nodes=55459969 evals=20379087 50move=0
endgame tablebase- probes=24907 hits=24629

mate in 7 moves.


White(1): Bc4
time used: 3:20
Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Crafty Problem with specifying tablebases path n o s p a m p l e a s e rec.games.chess.computer (Computer Chess) 2 January 26th 04 12:28 PM
Crafty SMP question John Cordes rec.games.chess.computer (Computer Chess) 0 September 17th 03 06:19 PM
Crafty Move List Christopher rec.games.chess.computer (Computer Chess) 2 July 25th 03 09:27 PM
crafty coding alternate line Christopher rec.games.chess.computer (Computer Chess) 1 July 25th 03 07:20 PM
Crafty Learning Robert Hyatt rec.games.chess.computer (Computer Chess) 0 July 9th 03 08:44 PM


All times are GMT +1. The time now is 11:12 AM.

Powered by vBulletin® Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Copyright 2004-2019 ChessBanter.
The comments are property of their posters.
 

About Us

"It's about Chess"

 

Copyright © 2017