Author: Bruce Moreland
Date: 14:25:21 12/18/02
Go up one level in this thread
On December 18, 2002 at 17:02:52, Omid David Tabibi wrote: >On December 18, 2002 at 16:59:49, Bruce Moreland wrote: > >>On December 18, 2002 at 16:26:36, Omid David Tabibi wrote: >>>Of course I have done my own tests, which confirmed std R=2's superiority over >>>std R=3. But I didn't publish them because they didn't indicate anything new, >>>just confirmed the previous published results. >> >>You published data that shows that R=3 is better than R=2. > >I can't see how someone can come to such a conclusion. (Are we talking about the >same article?!) Yes, again: You show in table 5 that you solve 849 problems through ply 10, with R=3, and 850 with R=2, also through ply 10. You show in table 4 that you get through ply 10 in 42% of the time with R=3. So you show two things: 1) Number of solution is almost identical. 2) R=3 is 2.4 times faster to finish. If you give me two versions that produce almost exactly the same solution set, and one of them is 2.4 times faster than the other one, it's very difficult to avoid the conclusion that the faster one is better. Tables 1 and 2 show almost the same thing. One solution worse, 2.2x faster. If you let me carefully specify a test for you to run, which makes use of conclusions from this data, I believe I can get your program to demonstrate that R=3 is significantly better than R=2. bruce
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.