Author: Ratko V Tomic
Date: 14:18:27 09/27/99
Go up one level in this thread
> Nothing is exactly the same as something else, this is obvious, but > what your theory and my joke have in common is that neither makes > sense. Having spent some years as theoretical physicist (including doing a thesis on paradoxes in QM probabilistic interpreetations), and then some as a consultant, among others in pattern recognition problems, I could only guess you missed the point. > If you look at the 6 best rated programs on a P200MMX, you will notice > that 5 of them (F532, Hiarcs, N99, F5 and Junior) are CB programs. First, this is a self-referential statement, i.e. you're defending the list by referring to the list's rating (or other lists, such as yours, calibrated by it). Second, Hiarcs 7 and Hiarcs 7.32 ought both to be tested on equal footing. Similarly for other multiple versions and products. Namely, if you have bought Hiarcs 7 or 6 and wonder whether you should buy Hiarcs 7.32 based on its strength, it would be worth knowing how do they both perform on the top hardware, not just the CB version. Third, in addition to different versions, there are also other programs which might have jumped at the top on K2-400, such as Rebel 10 and Century, as well as Mchess and Genius (in addition to CM and Tiger). Tripling or quadrupling the speed of hardware need not preserve the ranking from the slower hardware, especially when there is great overlap in rating uncertainty intervals from earlier tests. So, I would say there may be at least 9-10 top programs of interest to existent or potential users, and all of them should have gotten equal average hardware (some mix of K2-450 and P-200, if there wasn't enough fast machines for all programs and all games). It is absurd to rank them on the same list while systematically testing on highly disparate hardware, biased in favor of the 4 from a single manufacturer. It would be like racing cars, and the race organizers somehow decided that the 4 models from GM will be using high octane fuel, the rest only the low octane fuel, then putting the race results on the same rank list, even if you add a note what octane each car used. The GM can still crow that its 4 cars are at the top of that chart, even though that was decided well before the race even started, when the decision was made who gets high and who gets low octane fuel. Similarly, even a child could have told you, before SSDF played a single game for this cycle, that the 4 CB programs will the be on top of the list. If that's fair play to you, what would then be the other kind.
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.