21 Dec 04
Curious... assuming I have a chess program to which I can import a certain possition and it then goes off and plays against itself to determine a winner.
If this program reports to me the possible moves it made and the percentage of times it won vs lost with that variation, would this be cheating ?
21 Dec 04
Originally posted by pcaspianYes.
Curious... assuming I have a chess program to which I can import a certain possition and it then goes off and plays against itself to determine a winner.
If this program reports to me the possible moves it made and the percentage of times it won vs lost with that variation, would this be cheating ?
21 Dec 04
Originally posted by pcaspianYes, just cheating of a more sophisticated version that would be harder to detect, unlike mr.obvious JW. I hate to say it, but I suspect a good chunk of top rated players actually do this and they'll never get caught, but let's not get into that. Russ, mods, I suggest you delete this thread for safety, as this cheating strategy will just give players good ideas to cheat and avoid geting caught, I don't want this public.
Curious... assuming I have a chess program to which I can import a certain possition and it then goes off and plays against itself to determine a winner.
If this program reports to me the possible moves it made and the percentage of times it won vs lost with that variation, would this be cheating ?
21 Dec 04
You are proposing using chess engines to suggest moves. That is cheating unless the rules specifically permit it.
But if you have enough engines and processors to do this for every conceivable position, then save it all in a database that you refer to...
Well, then, we would be forced to play FRC for the last couple years that chess persists as a game. Fortunately, we are quite a long while from such a database (and it would probably take several terabytes just to store the thing).
21 Dec 04
Originally posted by rhbNo worries:
So eventually a computer bod is going to crack it, all conceivable moves will be 'mapped' and the game will be worked out?
Not sure I am keen on that. The beauty of the game for me is that I am never really quite sure if it will work or not!
The estimated number of atoms in the observable universe is 10^80; the estimated number of unique plausible chess games lasting 40 moves is 10^120.
Computers still have a way to go.
When RAM is measured in terabytes we might need to worry a little bit, but not before then.
22 Dec 04
Originally posted by SirLoseALotThe difference is between learning something from the computer for a specific game, and learning some theory through an example.
In theory,I would agree it's cheating.Yet,if one does what pcaspian says,after a game is completed,and use it for another game that has an identical or similar position,it would be ok.
What's the difference?
This is like the 'coaching' issue that mateulose raised against no1marauder. If it's not a game you're currently playing you cannot guarantee the exact same position will come up and that you can use the SPECIFIC information. Realistically (until computers have the required memory to deal with all permutations AND you print/save all the answers) you can only learn more general info.
22 Dec 04
1 edit
Originally posted by orfeoThe chance of the exact same position arising is not so small at all.If it's a complex middlegame,then yes,the chances are slim to none.But not so in an opening or endgame.Furthermore,there's no need to have the exact same position,just the part that matters will do.Example,say you have a position with an attack against the castled king,you don't find the correct continuation and win,draw or lose the game.You stick it in an engine,and it gives you the winning sequence.Few games later,you have the same attack,though in a totally different position,just the oppo's kingposition and your attacking pieces are the same,that can easily happen.You look up the previous game,and bam! you win.In both cases the engine has given you the moves.Yet,one case would be ok,but the other is considered blatant cheating.I don't get it,I fail to see the difference....
The difference is between learning something from the computer for a specific game, and learning some theory through an example.
This is like the 'coaching' issue that mateulose raised against no1marauder. If it's not a game you're cu ...[text shortened]... print/save all the answers) you can only learn more general info.
22 Dec 04
Originally posted by SirLoseALotWell, the sequences you're talking about are the kind of things people learn anyway, aren't they? Whether you use a computer, a book, or a more experienced player.
The chance of the exact same position arising is not so small at all.If it's a complex middlegame,then yes,the chances are slim to none.But not so in an opening or endgame.Furthermore,there's no need to have the exact same position,just the part that matters will do.Example,say you have a position with an attack against the castled king,you don't find ...[text shortened]... ok,but the other is considered blatant cheating.I don't get it,I fail to see the difference....
As I said in mateulose's thread, it's really the same as coaching in other sports, like tennis for example. The coach can go over the game in fine detail both before and after, just not during. The aim is that the player must play on their own DURING the game, but I don't know any sport that forbids you from learning before and after.
22 Dec 04
Originally posted by orfeoQuite right.And if we were talking OTB chess,I would be in total agreement with you.But we're not.This is corr chess,in corr chess we are allowed to look up things,we are allowed to learn during a game.The only difference in the example I gave is,the first game the player didn't have the data,or was too lazy to look for it,instead he had an engine generate the data,the second game,he had the data available and used it.If I think of it this way,I really see no difference.
Well, the sequences you're talking about are the kind of things people learn anyway, aren't they? Whether you use a computer, a book, or a more experienced player.
As I said in mateulose's thread, it's really the same as coaching in other sports, like tennis for example. The coach can go over the game in fine detail both before and after, just not dur ...[text shortened]... wn DURING the game, but I don't know any sport that forbids you from learning before and after.
Also,I had a database of enginegames,engine vs engine and human vs engine.Would I not be allowed to use that?After all,the moves are generated by an engine....
22 Dec 04
2 edits
Originally posted by SirLoseALotIn the first case, you are asking the computer to generate moves--clear cheating during a game, but not after.
Example,say you have a position with an attack against the castled king,you don't find the correct continuation and win,draw or lose the game.You stick it in an engine,and it gives you the winning sequence.Few games later,you have the ...[text shortened]... can easily happen.You look up the previous game,and bam! you win.
In the second case, you are consulting a book of sorts--perfectly acceptable in correspondence chess.
Classifying combinations is difficult, so if you even knew where to look, you were drawing on your store of chess knowledge. Everyone over 2000, and many of us that wallow at lower heights have a store of these combinational motifs in our heads. I try to teach the building blocks of these basic combinations to the kids I coach, while learning them myself from such texts as Polgar, Chess Training in 5334 Positions, Vukovic, The Art of Attack in Chess, and Renaud and Kahn, The Art of the Checkmate, among many others. In an early chapter of Renaud and Kahn, the authors explain Legall's Pseudo-Sacrifice--a mating attack from the only known game played by Philidor's famous teacher. I read this chapter a few years ago, and within a few months had twice--one month apart--used it against the same opponent (there were differences: one game was 9 moves, the other 11; neither ended in the exact position from Legall's game). Last month I claimed another victim.
I use engines for post-game analysis, to refine my opening repertoire, and to train against. I also regularly solve puzzles. The whole purpose of such training is to prepare for the second case. But, using engines to analyze a game in progress--never.
The difference is clear.
Now the dicta:
One advantage of slow time controls, and playing from home, is such use of books, and now databases. You can play a stronger game, but your opponents can as well. Sadly, most players treat sites like RHP--and the slow time controls--as another form of blitz. I know of players who play 100-200 games and more at a time on some sites. I've completed "correspondence-paced" games in three days and less. The quality of play in such cases is little different than playing 3 minute on FICS or ICC.
22 Dec 04
What ppl in RHP need to break from is the habit of claiming something is only "partial cheating", or "I am only a cheater sometimes". There is no partial cheating or a sometimes cheater, you are either a cheater or not, if you use an engine for just one move, YOU ARE CHEATING, you may never get caught and there would be no way to prove mathematicly you are cheating, but in your concious, you know it's wrong and you ARE a cheater, caught or not! Nuff said!
22 Dec 04
Originally posted by mateuloseI think you've missed the thread here. A serious question was asked about employing tactics in future games that were suggested by computers in analysis of previous games.
What ppl in RHP need to break from is the habit of claiming something is only "partial cheating", or "I am only a cheater sometimes". There is no partial cheating or a sometimes cheater, you are either a cheater or not, if you use an engine for just one move, YOU ARE CHEATING, you may never get caught and there would be no way to prove mathematicly you ...[text shortened]... ing, but in your concious, you know it's wrong and you ARE a cheater, caught or not! Nuff said!
Most of my 63 chess engines open with 1.e4.
Certainly I'm not cheating when I do the same.
Kramnik played on near auto-pilot against Leko's Marshall because he'd used computers to prepare. No one has accussed him of cheating; rather, many have pointed out that he should have let Fritz run a minute longer. Then, he might have anticipated the combination that Leko found on his own.