Could a player be detected if he starts using a chess program (using other computer) in the middle game or in the endgame?
Why do you ask? But yes I'd imagine so, particularly if they do it over multiple games. If you suspect someone of cheating, use lichess.org/report.
Disclaimer, I am not a Lichess moderator.
Disclaimer, I am not a Lichess moderator.
Yes Please use the lichess.org/report Don't waste your time on forum Posts.
Yes.
@JJCH_52 Yes, definitely. There are anti-cheat detection systems that can spot non-human moves based on the statistical analysis of millions of games, time-per-move and various concrete properties of networks and client hardware.
The assumption that some engine moves can be hidden amidst human moves is mistaken. They can be very obvious given the radical improbablity of certain players ever finding those moves.
The assumption that some engine moves can be hidden amidst human moves is mistaken. They can be very obvious given the radical improbablity of certain players ever finding those moves.
Yes when they leave the room and re enter while losing usually occurs at the end of the game
Many cheaters start their game with strange to ridiculous openings before they switch to engine moves. Probably they believe they are the most clever guys in the world and can trick the cheat detection with this marvelous idea.
ROFL
ROFL
@clutchnutz actually there are no "human moves" or "machine moves". Moves can be strong or weak, mostly ok, when you don't send pieces to the west, and even patzers like you and me sometimes find decent continuations. What a move makes really strong is the follow up and that you'll notice. Some time ago I got stomped by someone rated 300 below in a 3-0 on ICC, intitially a nice move followed by two sacs and mate, cool 7 moves combo played within 10 secs. I helped him to get his well deserved (C).
Definition: A *sequence of length n* refers to a consecutive chain of legal moves, where each player plays n moves.
Definition: A sequence of length n is *optimal* if there is no other sequence of length n with an equal or higher evaluation.
Definition: An optimal sequence of length n is *m-sharp* for 1<=m<n if any other length n sequence, which agrees with the first m terms of the optimal sequence, has much worse evaluation.
Fix n>=7. The times it's easiest to spot a cheater is when they play 3-sharp or 4-sharp moves. 3-sharp,4-sharp moves are the most 'machine-like', since for small evolutions of the board, they may seem suboptimal, and for any deviation from this long sequence (too long for most humans to calculate) they lead to a losing position. (Cheaters like this end up getting themselves banned, or lose the game - since either they follow the chain out and end up playing 10 stockfish moves, or they play something else, and it's bad)
Don't take my definitions too seriously :-).
Definition: A sequence of length n is *optimal* if there is no other sequence of length n with an equal or higher evaluation.
Definition: An optimal sequence of length n is *m-sharp* for 1<=m<n if any other length n sequence, which agrees with the first m terms of the optimal sequence, has much worse evaluation.
Fix n>=7. The times it's easiest to spot a cheater is when they play 3-sharp or 4-sharp moves. 3-sharp,4-sharp moves are the most 'machine-like', since for small evolutions of the board, they may seem suboptimal, and for any deviation from this long sequence (too long for most humans to calculate) they lead to a losing position. (Cheaters like this end up getting themselves banned, or lose the game - since either they follow the chain out and end up playing 10 stockfish moves, or they play something else, and it's bad)
Don't take my definitions too seriously :-).
This topic has been archived and can no longer be replied to.