Author: Sune Larsson
Date: 09:50:49 06/29/99
Go up one level in this thread
On June 29, 1999 at 12:19:17, James T. Walker wrote: >On June 29, 1999 at 10:46:35, Sune Larsson wrote: > >>I have followed the discussion of various Hash-Tablesettings with great >>interest. There seems to be an agreement(?) in the fact that Hiarcs 7.32 needs >>about 1/10 of the Hash-Tables needed by Fritz 5.32. So let´s see! >> >>PII-400 MhZ 128 MB RAM >> >>HashTables: Fritz 5.32 60MB Hiarcs 7.32 6 MB >> >>Books: Fritz 5.32 - F5-book Hiarcs 7.32 - Hiarcs-book >> >>Time: 60 minutes for the whole game >> >> >>RESULTS: Fritz 5.32-Hiarcs 7.32 13,5-6,5 >> >> >>F - H 1 1 ½ ½ ½ ½ 1 1 1 0 0 1 ½ 0 0 1 1 1 1 1 = 13,5-6,5 >> >> Comments ? >> >>Greetings >>Sune > >Hello Sune, >I have been getting similiar results all along. I have played 76 games at >Game/15 between Hiarcs7.32 vs Fritz 5.32 so far. The results are in favor of >Fritz by 46.5 to 29.5. They were played with auto232 with Fritz 5.32 on a >Pentium 333 with 64 M hash tables and Hiarcs was on a K6-2-400. I don't >remember the hash setting for Hiarcs but it was the minimum it would allow using >auto232. I think the GUI is set up for Fritz and the required hash tables for >auto232 play may be high for Hiarcs. I have also played 200 games at game/5 min >with the score in favor of Fritz 106.5 to 93.5. I played 200 game/5min vs >Junior 5.0 and Hiarcs 7.32. Junior 5.0 won 100.5 to 99.5. I have played 10 >"Long" games with Fritz leading so far 8 to 2. So, in my auto232 testing, >Hiarcs has not won a single time control. I am continuing the "Long" games now >playing 40/2hrs & 20/1hr with no sudden death. The required hash settings for >this time control is 44 Meg. Any of these games are available for examination >to anyone who request them. >Jim Walker And to your results I can add the following games Hiarcs 7.32-Fritz 5.32, autoplayed on a PII-400 MhZ 128MB RAM. H-F 1) 10'-games 44MB Hash each. Fritz' Powerbooks for both players. 5-5 2) 10'-games 44MB Hash each. Kingsgambitbook for both players. 6-4 3) 10'-games 44MB Hash each. Nunn-test. 10-10 4) 30'-games 44MB Hash each. Fritz' Powerbooks for both players. 16,5-13,5 I,m still confused concerning the best Hash-Table value for Hiarcs 7.32. The performance with 6 MB Hash-Tables was very poor. Or maybe there is something really fishy about Hiarcs' book! Sune
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.