Author: Sune Fischer
Date: 10:49:25 03/15/03
Go up one level in this thread
On March 15, 2003 at 12:06:50, Mark Young wrote: >On March 15, 2003 at 11:52:58, Uri Blass wrote: > >>On March 15, 2003 at 09:59:05, Kurt Utzinger wrote: >> >>>Comparison Fritz8 vs Shredder7 >>>Vergleich zwischen Fritz8 gegen Shredder7 >>>Comparaison entre Fritz8 et Shredder7 >>>Engine/Engine-Match, Time control 40'/40+40'/40+40' >>>Athlon 1.3/64 MB hash, ponder=off, 4-men TB's >>>Books/Bücher/Livres d'ouverture [write protected/schreibgeschützt] >>>******************************************************************* >>> >>>1) 100 x Fritz 8005 vs Shredder 7.00 [remis.ctg] result = 45,0-55,0 >>>2) 100 x Fritz 8005 vs Shredder 7.04 [remis.ctg] >>>3) 100 x Fritz 8005 vs Shredder 7.04 [f8-s7.ctg] result = 40,0-60,0 >>>4) 100 x Fritz 8005 vs Shredder 7.00 [f8-s7.ctg] >>>5) 100 x Fritz 80023 vs Shredder 7.00 [remis.ctg] >>>6) 100 x Fritz 80023 vs Shredder 7.04 [remis.ctg] result = 40,5-59,5 >>>7) 100 x Fritz 80023 vs Shredder 7.00 [f8-s7.ctg] >>>8) 100 x Fritz 80023 vs Shredder 7.04 [f8-s7.ctg] >> >>I see that the only version that I remember good comments of mark young about it >>(Fritz8.0.0.8) was not tested. >> >>I use Fritz8.0.0.8 and I decided to buy Fritz8 only after reading the comments >>of mark young. >> >>I dicided not to download updates for Fritz unless I read that the updates are >>better because I know that updates for Fritz often has new bugs and I do not >>want to spend time on testing engines when there is a big probability that they >>are worse. >> >>Uri > >After some bad results and testing with the newest fritz version. I will tell >you I was not impressed. Today I removed Fritz and reloaded the old and I think >better version of Fritz, 8.0.0.8. Fritz .23 plays about the same strength as >Deep Fritz 7 in my testing. I had much better results with .8. I will stay with >this version until I also see something better. I will keep .23 version of fritz >in a file so i can place it back into the engine file from time to time to do >more testing. I hope this is not the final version of Fritz 8. I would like to >see a better version then .8. :) Are you sure there is a larger difference than just bugfixes? If you look at the numbers: 8.0.0.8 and 8.0.0.23 it looks to me like these changes are very small being only fourth decimal changes, ie. probably merely bugfixes. I imagine Fritz is happily plodding along in development somewhere around version 8.5.4.21 pressing on to version 9.0.0.0. They might just go back to fix rapported bugs in the 8.0.0.x versions. Even if you played 100 games in selfplay you would probably not get a 50% score, so if they are close it's practicly impossible to tell. Personally I would trust the programmers in this case, their latest should be the best. -S.
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.