Author: enrico carrisco
Date: 16:29:55 12/13/03
Go up one level in this thread
On December 13, 2003 at 11:35:27, Omid David Tabibi wrote: >On December 13, 2003 at 11:11:19, Robert Hyatt wrote: > >>On December 13, 2003 at 05:31:25, Amir Ban wrote: >> >>>On December 12, 2003 at 21:11:15, Russell Reagan wrote: >>> >>>>On December 12, 2003 at 18:49:40, Amir Ban wrote: >>>> >>>>>There's no way to export a Chessbase book. >>>> >>>>You have Chessbase to thank for that. Let's be clear about who is to blame for >>>>you not being able to participate. It is not the fault of the volunteers who >>>>work hard to run the CCTs make good, reasonable decisions that make for a better >>>>event and promote progress. >>> >>>Well, if without Chessbase engines you'll have a better event and make progress, >>>I won't stand in your way. >>> >>>Amir >> >> >>What we have to wait for is enough interest in CCT that you can't afford to >>miss it. Also, for the record, the most demanded feature for chess engines >>by those that buy every one they can get their hands on is an automatic >>interface for the chess servers. I don't quite understand ignoring that >>demand. Or, for the first one to do it right, that level of extra sales. >> >>If the ICGA takes their responsibilities seriously, an automatic interface >>will eventually be mandated there although I personally prefer the CCT-sized >>event with 50+ participants rather than 16. > >A CCT style tournament can never turn into an official event, since you cannot >prevent any kind of cheating. I can run 5 engines on different computers, see >which analysis I like, and then force my engine to play that move by feeding the >move via a file it checks once a second. How are you going to prevent that? By >looking at the analysis I output?! I can force my engine to print a spurious PV >starting with the move I want it to play... > >The only reason why CCT tournaments are popular is that the stakes are not high. >Most programmers join CCT only to test their engine against others (speaking for >myself, I will enter a totally experimental and untested version of Falcon). But >when you give an official title to the winner, expect many (if not most) >participants to cheat in various degrees starting from "move now" to playing all >the moves as dictated by the operator. > >The physical presence of the programmers (or operators) is inevitable for an >official event, especially one giving the title "World Computer Chess Champion" >to the winner. That sad part is that the entire set of kibitzing rules came about for CCT5 after *ONE MANUALLY OPERATED* commercial entry that was being run by a very distant appointment of the author (who didn't even show up to the event) was giving the program longer thinking times and force-moving. During that same event, Fritz, Junior and Hiarcs were all auto-mated. What this means for those programs is that the native Chessbase GUI is in auto232 mode. Once in this mode, you can do very little to interfere with the engine's play -- other than exit auto232 mode. If you exit this mode, the winboard-engine adapter that is relaying the moves to ICC will no longer relay moves and it cannot continue or resume unless it replays the exact position from the beginning of the game (start position.) What all this means is that operator interference is very difficult -- nearly impossible. So what Omid has stated above is a good point -- It would be difficult but still possible to cheat and manipulate the event if someone wishes to do it -- However, not as easy as just changing a move with your mouse or changing the time on the clock for the program to think longer. Forcing programs to be automated was a very good idea and I don't remember anyone objecting. Going further with it and requiring the PV to be kibitzed by the engine before each move doesn't add any more security for those with cheating paranoia in my opinion. You've already "kept the honest people honest" with requiring the automation (which makes things tinker-proof as I described above.) Someone who still wishes to cheat or manipulate the event will rise to the occasion and do so. What the rule HAS done, however, is make the event off-limits to known and trusted participants like Amir and Mark, etc. unless they wish to re-write for winboard/UCI and dump their books totally. -elc.
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.