Author: Manfred Meiler
Date: 04:03:58 10/02/02
Go up one level in this thread
On October 02, 2002 at 03:41:46, Frank Quisinsky wrote: >On October 02, 2002 at 02:49:33, Jouni Uski wrote: > >>According to CSS WM position test it's specially good in positional and endgame >>play and relatively weaker in king attack. I think the last means same as >>tactics. But in my test suites it's also good tactically (e.g. it founds some >>well known mate in 18 and 30(!) quite fast). So may be the secret is, that it >>has no big weakness at all. Or what do You think? >> >>Jouni > >Hi Jouni, > >this program is very strong for king attack. > >A program must play the moves which are important for a king attack. Ruffian >make this. Not important are fantasy positions with a clear move. So I am not a >fan from test suits. > >In my opinion is the best test suite the LCTII test. >But look what Ruffian here make. Gandalf is over 120 ELO better compare to >Ruffian :-) > >An interview with Ruffian programmer Perola Valfridsson is in next days >available on Arena webpage (Ruffian is a partner program from Arena). It's one >of the questions from our interview team. > >Best >Frank Hello Frank, some days ago I've sent my excel sheet with the WM-Test results of Ruffian 1.0.0 and 117 other (versions of) engines (on AMD TB 1400 mhz) to Perola; maybe you're interessed on his answer ? "... Thanks for sending me this very interesting comparison of chess engines. I think you are right in your conclusion that Ruffian is weaker than some other engines when it comes to king attack. It is also my impression that Ruffian, compared to some other engines, is strong at endgame." But maybe you know the Ruffian engine better than the author himself ... You wrote about fantasy positions in WM-Test. I suppose you've never had a closer look at the 100 test positions; they're all out of games of the different (human) world chess champions. It's actually no problem for me that you are no "fan" (friend) of test suites; maybe the results of WM-Test ain't trustworthy for you. But it seems you have no respect for the (test) efforts of other computer chess friends. Manfred PS: The WM-Test results of Ruffian don't say that the engine is weak at king attack - but only that it's better in positional and endgame playing.
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.