Author: Chessfun
Date: 10:28:03 02/26/04
Go up one level in this thread
On February 26, 2004 at 08:22:51, Frank Quisinsky wrote: >On February 26, 2004 at 08:02:20, Thomas Mayer wrote: > >Hi Thomas, > >>On the other hand: You state in the mentioned posting that you believe that >>Ruffian 2.0.0 is about 50-60 Elos stronger then the version 23.06.2003 (which is the Leiden version, am I right ?) > >No! >Version 2.0.0 is clear stronger as the test version from 23.06.03 (not public >Ruffian version). > >>After the results of 2.0.0 were not as good as expected you said that it is not >>as strong an blitz. Later you said that it has problems with fisher time >>controls. And a bit later you said that maybe the Leiden version (23.06.2003 ?) >>is a bit stronger - because the v2.0.0 is - your words - in fact a beta version >>and was not tested very much. So what is correct now ? > >My comments are build from the results which I saw and my own. >I search the reason why users have different Ruffian results. >The most of the "bad" results I know from user which used ChessBase GUIs. >Within my first test was the Fisher time controls and the UCI Ruffian. >In the beta test some things in UCI mode are fixed by Per-Ola. Information can >be found in Arena Support Forum (longer beta test of Ruffian). > >>> I add my personal results in my forum and Arena webpages. I have no other >>> results. >> >>I think since the release of Ruffian 2.0.0 you should have now MANY more games - >>you even comment on many of the results - always with the same story that your >>own results show a difference... But for the conclusion in that posting - >>Ruffian 2.0.0 50-60 Elo stronger then Ruffian 23.06.2003 -> how many games did >>you have to claim that ? Just a question, not an insult of course ! > >I have played in the beta test time with Ruffian 23.06.03, later with the Leiden >version of Ruffian. Here I have played some games, but the most with 40 moves in >10 minutes. I public different tournaments with Ruffian versions in the last >summer on Arena webpages. I test Ruffian with private collected positions too. > >I have around 500 games with Ruffian 23.06. and around 400 games with Ruffian >Leiden. The most are 40/10, played under Arena Chess GUI. > >>You may remember the Gandalf-story... There you also said that it is one of the >>best if not the best engine at all. Gandalf was strong those days and still is - >>but there was a difference between dreams and reality. > >At this time my CCE tourney was running. >I believe the biggest tournament which I ever see in WWW. >18 months tournament time with games in 40 moves in 40 minutes. >Look in the SSDF and the Gandalf results. In the time of Gandalf are Fritz 6 and >Junior 6 available. The first versions of Fritz 7 are not so strong later are >Fritz stronger. You can see the different from Gandalf to Fritz 6 and Junior 6. >The same differences in my CCE tournament. The SSDF tested Gandalf on slower AMD >systems and Gandalf need time too ... you can see it now on the faster Athlon >1.2 GHz systems. > >You can see ... I give of every questions an answer. >Maybe we can make an interview :-)) > >>Besides that -> you might have read my complete posting: >>a) I pointed out that I still believe that Ruffian 2.0.0 is stronger then >>Ruffian 1.0.1 >>b) I even defend you that it is possible that your results seem to indicate that >>it is way better. > >Yes, I saw it! >At the moment I try to find out the problem! >With Shredder and many games I can say more and the results can be found in >Arena Event Forum with log files and so on. > >>Seems that you offend everybody who tries to defend you... not very kind... it >>seems that you still must realize that you are a businessman in computer chess >>now and must live with bad and good news. I believe that some of the older heros >>here like Ed or Ossi can tell you much about hits bellow the belt in that >>business. The SSDF-Result is of course not such a hit - it's simply reality. I >>hope that we will not see anotherone now you forces the SSDF to take an engine >>off the list. > >Not interesting what you know wrote! >I am user of chess software and computer chess is not my World of buisiness >after my "Erfahrungen" in the last years. > >The SSDF results are now reality, of course yes! >But more interesting is to search why the Ruffian results are different. >This is much more important for me. > >>Greets, Thomas >> >>P.S.: And believe me, the result of Ruffian 2.0.0 will get better - so far only >>~ 150 games are played and only 4 opponents. It might not jump 100 Elos ahead >>but I have no doubts that it will end up higher then Ruffian 1.0.1... > >Ruffian 2.0.0 is in fact around 75 ELO stronger as Ruffian 1.0.5 and within I >believe 100 ELO stronger as Ruffian 1.0.1. Do you know the results by Patrick >Buchmann, Alex Schmidt and much other persons in WWW. On the machine of Wilhelm >Hudetz the newer Ruffian won in front of Shredder. You can find a lot of such >results in WWW but bad results too. Now we have to test and to find out the >reason for it. I don't believe Ruffian 2.0.0 is anything near 100 points stronger than Ruffian 1.0.1. I don't honestly think anyone does, except maybe you. http://chessprogramming.org/cccsearch/ccc.php?art_id=342103 Sarah.
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.