Author: Rolf Tueschen
Date: 15:11:22 09/16/02
Go up one level in this thread
On September 16, 2002 at 16:00:29, Dann Corbit wrote: >Your inability to replicate results is in no way an indication that the results >are not genuine. > >1. There is a considerable amount of randomness intentionally built into chess >programs. Otherwise, once you found a winning line, you could beat them every >time using that line. >2. Every machine and every time control can give different results. On my >machine (for instance) Amy and ExChess seem to do much better than on other >people's equipment. >3. Inability to reproduce a measurement is not a disproof of the measurement. >4. It is possible to get wildly different results with similar setups because >of small, unforseen differences. For example, on a machine with less ram, you >might have the disk used as a replacement for memory, making operations >literally thousands of times slower than the same machine with more memory. > >You are welcome to question results. It is another matter to question someone's >integrity. Please stop it. Thanks for your explanations. Let me explain why I thought my discovery could be helpful? I thought that _such_ games with such apparent weakness in the opponent could not prove that Macheide Rebel had improved strength. That is the idea behind my posting and _not_ what you seem to insinuate the questioning of Thorsten's integrity. For me it was example for a game that couldn't support Thorsten's claim, the improvement in Macheide Rebel. I'm astonished that you take my postings with such bad interpretations. I didn't write a single word about integrity. But if the connotations or idioms look like I did I will honestly apologize here. But I doubt my English has such a bad content. Rolf Tueschen
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.