Originally posted by SirLoseALotinteresting. But once the accusation is made how do you know he's right? Surely ever accusation can't have a 'Poirit' ending where the accused makes it obvious he's guilty by trying to run. Do all the accusedown up???
Ahum,fritz7,old version.Don't boost the quality of your CM with false statements 😉😛
On the topic,I know a guy on ICC who's,so far,100% accurate at pointing out comp cheaters.He just goes through a game,rather quickly even and then he knows.This does not mean he can detect every cheater,but when he says you do,he's always been right,so far.He told ...[text shortened]... chess engine follows certain patterns?At least that's what I made out of his brief explanation.
Originally posted by belgianfreakWell,no,that is indeed a weak spot.The thing is,remember this is on ICC,we assume he's right because every guy he has ever pointed out did get a (C) label after some time.A (C) label means that ICC caught you using an engine.They have a team that examines games when suspicion of cheating is declared.And I think their software also detects engine use,but not everyone uses their software.
interesting. But once the accusation is made how do you know he's right? Surely ever accusation can't have a 'Poirit' ending where the accused makes it obvious he's guilty by trying to run. Do all the accusedown up???
So it's not 100%,but it are two independent sources catching the same player as a cheater.
SirLoseALot:
"Apparently a chess engine follows certain patterns?At least that's what I made out of his brief explanation."
Seems to me (with my limited understanding of the game) that in some games it would be easy to spot, and in others, hard. Look at the last decisive game of the X3D match- anyone who can beat Kasparov in a wide open tactical game, but play a closed position that poorly, is a computer. But it would be hard to spot this in a game that was just about hacking toward the enemy king- might just be a very strong player- you can only really recognize a computer by its mistakes. I guess I'm saying that any really strong player who regularly gets killed by anti-computer tactics is probably a computer.
That, I think is probably the best sign- a huge disparity between the quality of static and dynamic evaluations. Another good sign would be inept play in the ending until a tablebase position is reached, followed by perfect play.
Originally posted by mitiegeVery good points!Especially about the ending.never thought of that.
SirLoseALot:
"Apparently a chess engine follows certain patterns?At least that's what I made out of his brief explanation."
Seems to me (with my limited understanding of the game) that in some games it would be easy to spot, and in others, hard. Look at the last decisive game of the X3D match- anyone who can beat Kasparov in a wide open tactical gam ...[text shortened]... d be inept play in the ending until a tablebase position is reached, followed by perfect play.
The difference between tactical and positional is not such a good measure,IMO.There are lots of players who are very strong tactical but have no clue what to do when the game gets more positional.You know the kind,they live for tactical scrambles,play nothing but wild gambits etc...
Sir Lot.
Originally posted by SirLoseALot
[b]Very good points!Especially about the ending.never thought of that.
The difference between tactical and positional is not such a good measure,IMO.There are lots of players who are very strong tactical but have no clue what to do when the game gets more positional.You know the kind,they live for tactical scrambles,play nothing but wild gambits etc...
...[text shortened]... dynamic considerations- computers are much better than people at dynamics, and clueless at statics.
Looking more closely at this, looks like it was meant to trim the original, not the followup- sorry I was snippy. Still a bug, IMHO, but I write web forum software for a living, among other things- I am aware of the different issues that must be balanced- sorry if I sounded annoyed.
That said, this was the post I was making:
Sure, but there are limits- going from an effective FIDE rating of 2800 to less than 1400 is indicative (and I think Fritz' play was at least that bad in the game I refer to- a FIDE 1400 player (equivalent of 1700 or 1800 here, I would guess, not that FIDE rates 1400s) might have made a horrible move looking for counterplay there, but they would have looked for counterplay- dynamism, in essence)- static vs dynamic is not exactly the same as positional vs tactical, though there is a strong correlation. It is the unevenness of a computer's play that gives it away- you can't spot a computer from a game where it makes only good moves- might just be a very strong player. You have to look for static vs dynamic considerations- computers are much better than people at dynamics, and clueless at statics.
Originally posted by mitiegeAha!Got it loud and clear.Sounds like a decent theory to me.Seems like you know computers 🙂
Looking more closely at this, looks like it was meant to trim the original, not the followup- sorry I was snippy. Still a bug, IMHO, but I write web forum software for a living, among other things- I am aware of the different issues that must be balanced- sorry if I sounded annoyed.
That said, this was the post I was making:
Sure, but there are limits- ...[text shortened]... amic considerations- computers are much better than people at dynamics, and clueless at statics.
Guess that's normal since your job's writing software 🙂
I'm a computer dumbarse 😳
However- there is a sort of cheating that I think is pretty close to undetectable- which is computer-assisted play. At lower levels it might mean bringing in a machine only when in a tough spot. A look at the whole game wouldn't reveal a problem- after all, the patzer involved played badly enough to get into the tough spot... but below a certain level you can blow things pretty badly and still call in the cavalry.... and win a lost game. Who would accuse you of using an engine then?
At higher levels pairing a good player's positional sense with the machine's facility for calculation should be unbeatable- and also undetectable, unless you just assume that the top player is computer assisted.
- Joined
- 20 Oct 02
- Moves
- 610667
Originally posted by mwmillerSeems like a lot of work to increase a rating that more than likely, nobody really cares about except the cheater himself, which brings us to the old belief that when you cheat you are really cheating yourself....
It seems like this method of cheating is used regularly here at RHP, so take a closer look at this:
Cheater decides he has found a great new foolproof way to increase his rating. He will create a few Dummy players and have them lose to him after a few moves. This will allow Cheater's rating to climb quickly!
Now lets take a closer look at the way it ...[text shortened]... pen a few times, and I'm trying to comprehend how this makes any sense....
regards, Marc
Originally posted by Bobla45Exactly right in my opinion.
Seems like a lot of work to increase a rating that more than likely, nobody really cares about except the cheater himself, which brings us to the old belief that when you cheat you are really cheating yourself....
The methods of cheating are sometimes easily detected or in some cases, maybe not. But,...., what's the point, really?
The bottom line is the person who cheats doesn't really gain a thing. There is no money involved. If they get their rating up fairly high, one or two people may be somewhat impressed. Maybe they impress themself because they have managed to get away with cheating.
Even if nobody else is aware of it, the cheater knows they have a rating they do not deserve. I don't know why anyone would feel good about that. 😕
regards, Marc
