09 Nov 07
Originally posted by wormwoodI don’t follow… when I analyse a completed game with an engine, where does it say that my analysis lines should not extend many moves? The Fritz GUI has analysis features such as “Deep Position Analysis” and “Shootout” which can provide long lines.
the difference between that and following a complete engine game from a db, is that the db game will yield a high engine matchup rate, while improvements will only affect a couple of moves at best
This is a very grey area indeed. I bet the commercial databases from Chessbase (Megabase, etc.) contain games such as Kasparov vs Fritz, or Deep Blue, etc. Are we saying that all RHP players have removed such games from the databases they use with RHP?
By-the-way, these comments are not aimed at you personally. I’m just generally questioning exactly what the TOS is requiring or not.
09 Nov 07
1 edit
Originally posted by Varenkawhen you fritz a position yourself, you don't have the computational time to search anything but the few moves from your starting point, unless the variation is forced (but you don't need fritz in that case anyway). and there's not much chance the opponent will follow any unforced engine line for more than a couple moves at most.
I don’t follow… when I analyse a completed game with an engine, where does it say that my analysis lines should not extend many moves? The Fritz GUI has analysis features such as “Deep Position Analysis” and “Shootout” which can provide long lines.
This is a very grey area indeed. I bet the commercial databases from Chessbase (Megabase, etc.) contain g ...[text shortened]... med at you personally. I’m just generally questioning exactly what the TOS is requiring or not.
now, compare that to the situation where you'd have a fritz vs kramnik game in mega database 2006 for example. two CC players both have the same db, and will follow the engine game. one will score a 100% engine matchup for that entire game, not just for the couple of moves, while the other will get kramniks normal 76% matchup or something like that.
09 Nov 07
Originally posted by VarenkaI suppose you would be OK even without removing those games as long as you double-check that there's at least one human game in the line you're following.
This is a very grey area indeed. I bet the commercial databases from Chessbase (Megabase, etc.) contain games such as Kasparov vs Fritz, or Deep Blue, etc. Are we saying that all RHP players have removed such games from the databases they use with RHP?
09 Nov 07
Originally posted by VarenkaThere are loads of computer games (comp v comp and comp v human) on most databases, in fact far too many for my personal liking. However, I fail to see anything wrong at all in playing through games played by computers, if that's what interests you (it doesn't me). Or have I misunderstood something?
I don’t follow… when I analyse a completed game with an engine, where does it say that my analysis lines should not extend many moves? The Fritz GUI has analysis features such as “Deep Position Analysis” and “Shootout” which can provide long lines.
This is a very grey area indeed. I bet the commercial databases from Chessbase (Megabase, etc.) contain g ...[text shortened]... med at you personally. I’m just generally questioning exactly what the TOS is requiring or not.
10 Nov 07
Originally posted by incandenzaYes, but what if the stats for a given move include the wins, etc. from engine games and the user chooses a move using the stats (as many do)?
I suppose you would be OK even without removing those games as long as you double-check that there's at least one human game in the line you're following.
10 Nov 07
Originally posted by Northern LadIt becomes relevant when a database is used for choosing RHP moves. The TOS is specifying that the database should consist of human vs human games only.
There are loads of computer games (comp v comp and comp v human) on most databases, in fact far too many for my personal liking. However, I fail to see anything wrong at all in playing through games played by computers, if that's what interests you (it doesn't me). Or have I misunderstood something?
10 Nov 07
Originally posted by VarenkaWell, the fraction of engine games in one of these large databases is probably vanishingly small (0.01%? 0.001%?) It would be pretty hard to find a case where the statistics would be skewed by a noticeable amount.
Yes, but what if the stats for a given move include the wins, etc. from engine games and the user chooses a move using the stats (as many do)?
But you're right, I suppose technically they should be removed. I would say it is a minor violation of the 'letter' of the rule, but not the 'spirit' of the rule.
I wonder if anyone who uses a large database (like the Mega database or whatever) actually bothers to remove the engine games?
10 Nov 07
1 edit
Originally posted by myteamtrulystinksYes... I've played several games that follow the 'book' all the way to a draw, mostly against higher rated players though so I'm okay with it 🙂.
My buddy and I (we are both 1500 players) were always resistant to the concept of using a data base. We felt it was cheating, but a player convinced my friend that his play would be boosted by it. So slowly he started using it and it he said it helped him. However, he just played a game and showed me how the data based helped him. But it sort o ...[text shortened]... this sort of happen from time to time when you use a data bases or is this some sort of overuse?
edit - Game 4192051 - Here's my most recent game where we followed the book all the way to the end. A game between some GM who I never heard of before (Van Der something) vs. Kasparov followed this exact line and it seemed that by the time my opponent and I figured out we were following the same line, it was too late to deviate safely! Pretty funny actually.