Author: Brian Kostick
Date: 13:28:42 01/13/01
Go up one level in this thread
On January 13, 2001 at 13:51:47, Hermano Ecuadoriano wrote: >On January 13, 2001 at 13:03:35, Brian Kostick wrote: > >>On January 10, 2001 at 21:14:48, Robert Hyatt wrote: >> >>relevant snip (refering to 10,000 rpm code post): >>>Are there any source changes I should make to make this an easier thing >>>to do? >> >> I'm wondering if wcrafty-18.01.exe (or whatever is current) might become >>available at your ftp site?? I would like to run "bench" on it when it is >>compiled with the same compiler as wcrafty-17.14.exe. >> >>Here (733-PIII)I get a wcrafty-17.14 bench of 122sec. and a Crafty1.exe(Corbit) >>bench of 150sec. Is it realistic that version 18 would be slower? At any rate I >>think side by side test would be nice to run before you bothered with source >>changes. If this is already done then my post is in vain and I'm sorry to bother >>you. >> >> I can compile with the Intel 5.0 compiler (trial run) with minimal changes to >>the makefile.nt. I don't have the optimizing MS compiler if that is what was >>used to build wcrafty-17.14.exe. >> >>Thanks for any info/insight that you can afford, Br5an > >PII-300 48M hash, 5M hashp > >version, nodes, n/s, time > >wcrafty-18.1 56199841 172922 325 >CRAFTY1 56199841 153133 367 >wcrafty-17.14 52518631 170515 308 Thank you for posting your timings. If I interpret correctly you are confirming my suspicions. Might I ask where you found wcrafty-18.1.exe?? (or did you build it?) I currently do not see it at Mr. Hyatt's ftp location in the 18 directory as was common with previous versions. Br5an
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.