Author: Uri Blass
Date: 00:22:29 12/07/02
Go up one level in this thread
On December 06, 2002 at 18:08:52, Bertil Eklund wrote: >On December 06, 2002 at 10:36:08, Mike Hood wrote: > >>Rex, I wanted this to be a Ruffian thread, not a Chessmaster thread, but I have >>to answer what you've said. The non-testing of Chessmaster and Ruffian are two >>completely different matters. Chessmaster doesn't offer an an autoplay function, >>for whatever reason, so it would mean a lot of hard work for the SSDF guys to >>test it. The King engine runs under Winboard, but since this isn't the way >>Chessmaster was intended to be run, UbiSoft could justifiably shout "UNFAIR!" if >>it were tested in this manner. >> >>Ruffian, on the other hand, is a UCI-Engine intended to be run as UCI. Neither >>the fact that it's new nor that it's freeware are relevant. Other free programs >>are tested by the SSDF (such as Crafty and Comet). Other progams are tested by >>the SSDF that are newer than Ruffian (such as Chess Tiger 15). >> >>It comes down to the fact that with the large number of chess programs >>available, the SSDF has to decide how to allocate their limited resources. The >>SSDF would be fully justified in banning from their list all programs without >>autoplayers. This might annoy the fans of strong programs like Chessmaster, but >>it's up to the owners of Chessmaster (UbiSoft?) to provide an autoplayer. I'm >>certain they already have one in their offices, but they've decided not to sell >>it. Write to them and complain! >> >>As for Ruffian... it's the SSDF's call. Are there enough reasons to justify >>testing it as one of the many Winboard and UCI engines? I would like to see how >>it fares against the big commercial programs, because there have been posts in >>this forum claiming it is "better than Fritz 7". Although I suspect that these >>claims are exaggerated, it would be good to see Ruffian being tested. But as I >>said, it's up to the SSDF, so, to rephrase the title of this thread, "Why hasn't >>Ruffian been tested?" >> >>On December 06, 2002 at 07:16:57, Rex wrote: >> >>>I agree.. This is yet another example of why I question this SSDF. Chessmaster >>>9000 can run inside winboard (ok it doesnt have its own opening book but make >>>both engines use the nunn opening book!!) and Ruffian is not included. These >>>two issues should be questioned. I am beginning to wonder if buying chess >>>programs are worth my damn money anymore not knowing not ALL programs are being >>>tested. >>> >>> >>> >>> >>>On December 06, 2002 at 05:05:03, Mike Hood wrote: >>> >>>>Is there any particular reason why Ruffian (allegedly the strongest freeware UCI >>>>engine) hasn't yet been tested? Will it be included in the next SSDF list? > >Hi! > >Of course we try to include Ruffian in the next list. I guess, if it compares to >well with the best programs, people starts to questioning the results because it >is a Swedish program! I have played a few houndred speed games with Ruffian and >at least with the Shredder-book it is an excellent program, it has only one >weakness (well covered with the big book of Shredder) the opening phase, where >it plays like a program from the early eighties. > >Bertil I do not believe that there is an engine with only one weakness. I do not know about the opening phase weakness and maybe it is better to give examples to see if the weakness does not disappear at 40/2 hours but I believe that bad time management is one of the weaknesses of ruffian. It can be more clear if you try 300 moves/5 minutes when ruffian use less time than 5 minutes/game. It proves that ruffian has not a good formula how to use time(the fact that this mistake is smaller in 120 minutes/40 moves+120 minutes/40 moves changes nothing because it is still a mistake to divide the time by the number of movesto the next time control to decide about the target time (something that ruffian seems to do) and the fact that many engines do this mistake does not change the fact that this is a weakness. I also believe that ruffian has weaknesses in the evaluation(every program has and ruiffian is no exception) I played in the past few matchs between movei and ruffian(movei with no book and always change it's first move and ruffian with it's original book) I remember a game when ruffian lost against movei because of overevaluating king safety(movei has no king safety evaluation). Movei lost the last 2 matches 37.5-2.5(7 minutes per game) and 38.5-1.5 few weeks ago (the 38.5-1.5 happens at the longer time control 20 minutes/game if I remember correctly and the only game that ruffian lost was because of overevaluating king safety and not because of bad opening). Uri
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.