Author: Robert Hyatt
Date: 06:44:35 02/24/98
Go up one level in this thread
On February 24, 1998 at 03:51:59, Peter Herttrich wrote: >On February 23, 1998 at 14:58:09, Moritz Berger wrote: > >>On February 23, 1998 at 14:45:49, Thorsten Czub wrote: >>>The problem is : WHEN fritz5 is that strong, why didn't we found out >>>about ? >>>Why didn't you tell me before. Or Peter Schreiner. Or any other guy >>>HAVING FRITZ at home ? >> >> >>I told you a long time ago that Fritz 4 was strong and belonged among >>the top programs. > >Here i have other experience and not only i. >Fritz4 has played about 20 Games against Crafty 14.5-14.10. >And has won NO ONE GAME. The result was about 70% for crafty. >All others are draws. Environment: K6/200 with 48MByte Hash for >Fritz, K6/200 48MByte Hash for crafty (of course two machines). >And under 40/120 tournament condition! >I was really frustrated about F4, it plays so much junk. >Thank god i have payed only about $30 (50.-Deutsche Mark), it was the >special edition with the 4.01 engine. >Dont tell me F4 is strong! Again, I suggest not using this with too much confidence. I saw similar results against Fritz 5. And I have no explanation as to why this was true (don't know if it is still true now.) But testing against a single program is not a good idea. Fritz and Crafty are probably reasonable similar in search techniques, although I can't guess about his evaluation. But if the most important part of two programs is the same, then other minor differences can add up to significant winning/losing margins even though the two programs might be almost identical in skill when playing human or other program opponents. This can be quite maddening. I remember the first example I saw where the Deep Thought guys thought their singular extensions were worth almost 200 rating points based on comparing DT/with to DT/without. That tends to exaggerate small differences. I personally like the way F5 plays. It seems to like active positions and plays aggressive moves. Whether it is fast and smart or fast and dumb, it is *certainly* fast and "good." :) >If i remember right F4 played similar in Jakarta. > >cheerio >peter actually Jakarta was a debacle, but Frans showed up with a new version that played about like Crafty played in Paris. Things that look good in testing can fail miserably against decent competition (different competition). Certainly happened to me...
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.