Author: Enrique Irazoqui
Date: 04:17:22 03/10/98
Go up one level in this thread
On March 10, 1998 at 05:27:23, Ossie Weiner wrote: Dear Ossi, During the last few days we had this discussion in an email group. It's getting repetitive, but let's try again. >For many years the SSDF rating list has been a trusted source of >information >for computer chess enthusiasts. It used to be the basic principle of >this >non-commercial organisation that chess computers and chess programs were >tested only in exactly the same form which was also available to the >public. >This straightforward principle made the SSDF ratings valuable for all >interested parties. > >For the first time now the SSDF has allowed a company called ChessBase >GmbH >to supply a special hardware/software configuration which is not >commercially available. That means special privileges have been given to >ChessBase in comparison to other chess software suppliers. It has always >been usual to give equal conditions to everybody, but now we find this >principle of fairness violated in the latest SSDF rating list from >22.02.1998. This is simply untrue. The first time SSFD tested a non-commercial program was a couple of years ago with Gemius 4 for DOS, not available to the public and released to the SSDF precisely by you. It is you who established the precedent. In any case, the engine in the autoplaying version of Fritz 5 is identical to the commercial version released on the Junior CD-ROM. >The following points are making a suspicious impression: > >a. It's known that ChessBase had autoplayers available latest in >September 1997, whereas the SSDF received it only in beginning 1998. >What >has ChessBase been working on in these 2 - 3 months? An updated engine, commercially available. >b. Unlike all other chess programs Fritz 5 has been tested with a >special Powerbook loaded onto the hard disk for speed increase. These >Powerbooks are normally not supplied with Fritz 5. The Powerbook is available commercially. Nothing "special" about it. Out of the box, standard Powerbook. >c. Fritz 5 is being tested with endgame databases. Has this also >been >the case with other programs such as MChess Pro 7.1 and Shredder 2.0? >d. We have been informed that the exe-file used in these tests is >different from the commercial Fritz 5 exe-file. Why is the SSDF >tolerating >this? Again, the chess engine is the same. Whether or not the exe file contains the autoplaying driver it's irrelevant. >e. Fritz 5 is not being tested with a standard auto232 interface, >but >with a very special hardware/ software configuration supplied by >ChessBase. >This setup opens the door to various manipulations such a special tuning >to >different opponents. The possibilities of manipulating when autoplaying are open to all programmers. It does not depend on using the DOS auto232. Shredder 2 and Nimzo98 have also their own autoplaying driver. > It also requires a min. RAM size of 64 MB, a >privilege >nobody else has demanded or been granted up to know. >f. Some experts have calculated only the handtested games of Fritz >5 >where the achieved rating is almost 200 Elo points lower. Only >coincidence? Worse: it is false. Look at the results posted by Eric Hallsworth on his Selective" magazine. All games played manually, and Fritz 5 is in the group of top programs. No difference between his manual results and my autoplayed ones. Ossi: you state false accusations and at the same time you have been repeatedly refusing to provide the proof I asked you. It is very simple. SSDF posted their games autoplayed by Fritz 5. The score includes time per move, evaluation and depth. Any owner of the updated Fritz 5 can verify that both versions, the commercial and the autoplaying, play the same move, with same evaluation under the same time. I asked you publicly to show us one game, one would be enough, where these two versions play differently. Instead, you prefer to keep rambling about possible cheats. One more point: the Powerbook is the standard, commercial CD-ROM anybody can buy. It does not include cooks. I have over 700 games autoplayed by Fritz 5. In no game Fritz left the book with an advantage significant enough to talk about killer lines. This is also possible to check in the scores of the games played by the SSDF. >There is the big danger that the SSDF rating list will become worthless >in >the future, unless the SSDF is returning to their basic principles: > > All chess programs have to be tested in the commercially >available > form WITHOUT tolerating special requests of any manufacturers. > >That means that the Fritz 5 special version has to be removed >immediately >from the rating list. Only games may be counted which have been tested >by >hand, not by the mysterious autoplayer configuration. As soon as a >sufficient number of games has been tested in a regular manner the >results >shall be published again. Of course also autoplayer games are >acceptable, >but only if it's done in a commercially available form, to which every >computer chess friend in the world has access. We all have access to the games and the scores. With them you can check instead of speculating and posting false accusations. Enrique >Of course we can't exclude that Fritz 5 is really that strong and it >will >achieve first place on the rating list also with regular testing >methods. In >such case nobody will object to such a well-deserved result. But >currently >the possibility can't be excluded that this program may drop by as much >as >100 Elo points or even more, and for that reason it's also in the best >interest of ChessBase GmbH to stop all negative rumours and present >their >program to a FAIR TEST. > > >Signed by: > Chrilly Donninger > Richard Lang > Stefan Meyer-Kahlen
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.