Wow... looks like I have to be careful not to improve too quickly. Otherwise I can't play on here no more.... hehehe. But seriously, Dragon Fire, I haven't played for some years now... and then only recently came back to the game. I used to play OTB tournaments and spent long hours at night hovering over a chessboard. I've been missing some obvious moves on here in the beginning. But now am improving back again. Am also reading up quite a bit this lately. Still a lot more theories to acquire. Of course I know quite a lot of basic stuff. I guess one never really forget everything he's learned before.
I am quite serious at this game, but have not actually spent the time to analyse my games using a computer program. Perhaps I must get one soon if I want to improve further. In the mean time, it would be good if someone can explain on how the moderators can detect cheats on here. Or are they kicking people out solely based on suspicion?
I'm in the same boat.
Played when I was young a lot, had tuition, played in a local club for a few years against old men when I was 10 to 12 years old.
Then a mate suggested this site & I started playing again for the first time in years.
A few hit & miss games (& playing whilst drinking wine) meant mixed results.
I got my enthusiasm back though. I bought a few books mostly about openings, tactics & strategy & my rating has steadily increased to over 1500.
It is essentially as though you are starting from scratch after so long out of it - I'm 34.
I still make naiive moves & still blunder & only use a very limited number of openings - King's Gambit, Ruy Lopez, Morra Gambit, Grand Prix Attack as white & Sicilian against 1.e4 & the Slav against QG.
I think there must be a way of telling if someone is blatantly cheating, such as they are rated 2400 & don't play OTB chess etc or don't discuss their games or moves.
They may make many moves in many different complex games against highly rated opponents. That would flag a cheat to me.
Don't computers play differently than humans? - they focus on small tactical improvements in positions, but not the overall strategy or position.
If you are not using a chess engine to make your moves the chances of you being banned from RHP on 3(b) are infinitely small, regardless of what your rating graph looks like, or how much your game has improved.
As game moderators, we have to provide overwhelming evidence of engine use to the Site Admins before they will take action. There is no way we can do that if you are a human being and playing your own game. But once you switch that engine on and start playing like a computer you run the very real risk of having your account terminated.
There is a very wide gulf between strong human play and engine play. Even at the very highest levels, humans do not play like computers.
Originally posted by ckoh1965You are allowed to use databases of past games and opening books. Once the game is over you can give it to an engine to crunch to see what you might have missed. What is cheating is to use an engine to analyse the game before it is over to help you find moves. Also you cannot use endgame tablebases.
I am quite serious at this game, but have not actually spent the time to analyse my games using a computer program. Perhaps I must get one soon if I want to improve further.
In the mean time, it would be good if someone can explain on how the moderators can detect cheats on here.
Or are they kicking people out solely based on suspicion?
If someone makes a complaint against you the moderators analyse your previously played games and look for correllations between what you did and what an engine would do. It's not a question of playing very accurately, engines play differently to humans. If the match is very high, over 90% (I don't know the threshold the mods use, that figure was picked out of the air) then they'll conclude you have been using an engine to make your moves and you'll be banned. Even grandmasters don't get match up rates above 75% (or so).
No, they are careful not to brand the innocent as cheats, as if they did there would be a rapid exodus of players - a few people left the site in protest at the first wave of bannings and those banned were clearly guilty. It is not in the site owners' interests to remove players from the site without good justification as they don't get subscriptions off them if they do.
Originally posted by DeepThoughtThere are a few games I play when I face an innovation or simply an inferior line not in my books. When that happens I will use Fritz to analyse the game afterwards and note this analysis in my "books" and in a database for subsequent use. This is not cheating but in such continuations my moves will surely have an improbably high correlation with Fritz and in exceptional cases I will find and use that non-human move.
You are allowed to use databases of past games and opening books. Once the game is over you can give it to an engine to crunch to see what you might have missed. What is cheating is to use an engine to analyse the game before it is over to help you find moves. Also you cannot use endgame tablebases.
Perhaps because this happens very rarely and even then only for a few moves at the commencement of that new line it would not reflect in my overall performance but I make the point that there could be very valid reasons why the occasional move does appear non human.
Of course this could only happen in the opening when we are still in (or just out of) book - it could not occur later in the game where it would be unlikely I ever faced a previously occuring position. In fact generally in openings when databases and books can be used matches will presumably be consistently high. Having said that in one of my favorite openings, The Kings Gambit, Fritz does not like whites position at all and keeps trying to win that pawn back too early and gets the strategy totally wrong. It takes it a long time to decide white is equal let alone better. Gambits don't seem to go down well with computers.
Originally posted by DeepThoughtBased on what the mods and other strong players have posted in the past, there is more to it than that. I am sure it would have to be an analysis over several games, and situations where there are strong engine type moves.
If the match is very high, over 90% (I don't know the threshold the mods use, that figure was picked out of the air) then they'll conclude you have been using an engine to make your moves and you'll be banned. Even grandmasters don't get match up rates above 75% (or so).
There are occasions where a human can have a high match up to an engine even after database moves have been left (most likely short games).
It wouldn't be good banning people who go over a certain threshold, although this is probably a good starting point to initiate further analysis.
It becomes most relevant when a cheater is playing a strong player. Against weaker opponents opportunities would exist to deliberately make inferior moves that still maintain the winning advantage but once strong opposition is faced the cheater would need to select the top engine moves consistently.
Not having the skill to tell the differance between an engine move and a human move (due to not being good enough to play human moves) such a player would make obvious engine moves than no strong human player would make.
Cheaters need to beware. Once they play strong opponents they will be caught and if they don't play strong opponents they will attract suspicion as any strong human player would want to play the best possible opponents.
Also, and I hate to say this, strong players devote less time and effort to games against weaker opposition. They should not do this because sometimes you come unstuck but being human we do. Their moves against such players will therefore be "less complex". They will not waste time going for that complicated tactical variation that takes ages to calculate but will go for that simple and obvious win. A comparison of games against strong and weak players will reveal differances in a human game that is not evident in an engine.
Originally posted by lauseyYou're right. I was more concerned with clarifying what you can and cannot use an engine for. The statement about engine use detection couldn't be complete because the games mods don't give all the details on how they do their work for obvious reasons. I imagine that some account is taken of how likely a human and an engine is to agree. Although frankly if someone is averaging a 90% matchup rate over 100 games then it really is no accident.
Based on what the mods and other strong players have posted in the past, there is more to it than that. I am sure it would have to be an analysis over several games, and situations where there are strong engine type moves.
There are occasions where a human can have a high match up to an engine even after database moves have been left (most likely short gam ...[text shortened]... ertain threshold, although this is probably a good starting point to initiate further analysis.
no points, damn it. I was holding my own with black against gauravv in
Game 2491112
with 0.02 (fritz 9), and probably killing him in the other game
Game 2491111
I already had the taste of sweet victory in my mouth! but all in all, the hellfire clan (avg rating in challenge around 1800) drew against faith hope love (avg in challenge around 2100), and I'll take that for a victory!