Originally posted by Pigface1I don't see how that could be implemented here, this is a website, not a program. There would have to be all manner of very intrusive javascripts in the RHP webpages. And they would be completely useless as someone could get round them by closing their browser, running the engine to find a new move, closing the engine and restarting the browser.
Maybe that could be implimented here, it would drastically reduce the chances of people cheating, even though you could work around it and just have one open at a time, many who did cheat might not want to bother with that
Originally posted by Nordlyswhat I've read about ICC cheat detection makes me believe they do exactly what game mods used to do here: receive a complaint, and do a statistical analysis on past games.
It's probably also easier to detect cheating in real-time chess as opposed to correspondence chess. So what would be really interesting would be other CC sites that handle this problem better than RHP.
I also really don't believe having an engine running while on playchess has any meaning. because 1) people might want to play while the computer is busy analysing previous games, 2) they could easily use any non-chessbase engine, and 3) it completely ignores all other possible kinds of cheating. it just wouldn't work. I think it's a rumour.
Originally posted by crazycolBut I thought we already have the ability to report people for suspected cheating...the only difference with your system is that a given number of complaints would automatically trigger an investigation. I don't see the benefit in this as it's not just the quantity of complaints but also the quality that a moderation team might wish to consider. One complaint from an experienced player coupled with strong evidence arrives - why would they wait until more complaints come in.
True - they could, but I guess that's the benefit of having a decent number of Yes marked games to trigger the 'suspect' rating. Plus it would mean an investigation - not an immediate guilty verdict. It may end up being counter productive for engine users to falsely feedback Yes 9/10 ratings as it would invariable ensure analysis of games involving their ow ...[text shortened]... errogateable data and ensure players feel like they are contributing in an effective manner.
Also with my rating around 1500ish and over 200 games played I don't think I've ever played a cheat and I wouldn't be able to spot one if I did.
It wouldn't surprise me if less than 5% of users actually visit the chess forum & maybe 85% of people on this site are below 1500, how many people do you think are actually directly affected by the engine use problem?maybe 85% of people on this site are below 1500, how many people do you think are actually directly affected by the engine use problem?
To the vast majority this is simply an issue that has no relevance.
If people are that bothered they should try to improve their online blitz play or concentrate on OTB.[/b]
Perhaps not directly affected but the top players are to my mind crucial to the success of the site.
What would the site be like if most human players over 1700 abandoned the site due to unconstrained engine use.
This would damage the community and so makes the issue of cheating relevant to everyone even those not directly affected.
I do think it's an option to just mark the username with a (c) and remove them from the rating pool rather than banning. On ICC for example you can log on and play an unrated game against a clearly identified computer if you wish and some newbies might prefer this option for a game or two when they start. You can also practice endings against a computer and so on.
Originally posted by SquelchbelchI apologize for going off on you. 😞
Twisting my words.
I said that to the vast majority of members the engine use issue is of little concern.
I reckon the average rating here is proabably around 1300. So how many in the massive 1200-1400 section do you think are using Fritz to enhance their performance?
How many of these guys (who presumably mostly play people of similar ratings) ar aps a 1900+ is equally useless at understanding what goes on in the minds of us mere mortals?
What I should have said is that there are people who cheat at all levels. I usually put my losses into fritz afterwards to see where I could've played better. It's amazing how a 1500 rated player, who was playing very badly, suddenly starts finding brilliant moves that match fritz's play from there on.
I should have said that if people don't complain and say it's not a problem then it will get worse.
I have some ideas on how to reduce the problem and have submitted them to the site admins. My suggestions would involve bringing back the mod team though.
Originally posted by wormwoodI agree.
I also really don't believe having an engine running while on playchess has any meaning. because 1) people might want to play while the computer is busy analysing previous games, 2) they could easily use any non-chessbase engine, and 3) it completely ignores all other possible kinds of cheating. it just wouldn't work..
There is no way for a client app to know for sure that the same PC is not running a chess engine. Maybe some approches prevent the most straightforward cheaters, but nothing beyond that.
Originally posted by MahoutNo problem, it was just an idea. One that I admit is not flawless 🙂
But I thought we already have the ability to report people for suspected cheating...the only difference with your system is that a given number of complaints would automatically trigger an investigation. I don't see the benefit in this as it's not just the quantity of complaints but also the quality that a moderation team might wish to consider. One complain ...[text shortened]... played I don't think I've ever played a cheat and I wouldn't be able to spot one if I did.
Judging by the consensus of opinion I'd guess that most would like something to be done regarding either catching or deterring cheats. It's just they don't agree on methods or viability.
I prefer to think of positive ways to improve a situation rather than to negatively reinforce it.
Originally posted by MahoutMy initial post mentioned Ebay as a comparison. The idea was that the system would maintain its own trust levels without site admins getting involved. Whether this would work for an online chess site or not, I don't know.
I don't see the benefit in this as it's not just the quantity of complaints but also the quality that a moderation team might wish to consider.
For example, just as we have tournaments with rating criteria, we could have "reputation" criteria. If too many people have suspected you as cheating, you can't enter. All automatic; no manual intervention. Of course, just as Ebay have to try to make sure that reputations don't get skewed unfairly (up or down), so would RHP. Maybe there would be unstoppable abuse mechanisms; again, I'm not sure.
Originally posted by Varenkawhat if 1 higher rated player called out someone for cheating in the forums? didnt say any details.. just said cheating.. wouldnt he get super low just due to that?
My initial post mentioned Ebay as a comparison. The idea was that the system would maintain its own trust levels without site admins getting involved. Whether this would work for an online chess site or not, I don't know.
For example, just as we have tournaments with rating criteria, we could have "reputation" criteria. If too many people have suspecte ...[text shortened]... n), so would RHP. Maybe there would be unstoppable abuse mechanisms; again, I'm not sure.
Originally posted by VarenkaWhat I would be worried about with such a system is that it might turn into a popularity contest. People are more prone to suspect or accuse someone of cheating if they dislike the person. Some of the people who have been banned for engine use here probably wouldn't have got into any problems if it had been the public opinion that counted, because they were popular and didn't have overly suspicious rating graphs; on the other hand, there might be unpopular players who'd be accused despite being innocent. And then there are those players who accuse others of cheating just because they were beaten - I have seen quite a few accusations of that sort in the fora. I also wouldn't exclude the possibility of a group ganging up against a player.
My initial post mentioned Ebay as a comparison. The idea was that the system would maintain its own trust levels without site admins getting involved. Whether this would work for an online chess site or not, I don't know.
For example, just as we have tournaments with rating criteria, we could have "reputation" criteria. If too many people have suspecte n), so would RHP. Maybe there would be unstoppable abuse mechanisms; again, I'm not sure.
Personally I don't think I am able to tell whether someone is cheating or not apart from some extremely obvious cases, and in a system like you suggest, I probably would refrain from reporting someone most of the time, as I would know that it would affect their reputation even if I am wrong. So I prefer a system like the game mod system where you can report someone in the knowledge that it won't have any negative consequences for that person if they are innocent.
Originally posted by wormwoodWell, I don't know about Playchess, but both of you are mistaken to think that a determination can't be made if the person is using a chess engine while playing online at some sites. I know of one site that simply asks the Windows operating system which program is "in priority"? Which is the same meaning as 'tell me the name of the program running'... Ya-see, the Windows operating system will gladly give this information out if requested from another program. All that has to be done by the interrogator program is ask repeatedly every few hundred milliseconds in order to establish a time table.
I also really don't believe having an engine running while on playchess has any meaning.
Originally posted by Varenka
I agree.
There is no way for a client app to know for sure that the same PC is not running a chess engine. Maybe some approches prevent the most straightforward cheaters, but nothing beyond that.
Sooo, if a program asks the operating system for this simple information the response back can look something like this:
fritz 9.0, 3.5 sec
playchess, 5.1 sec
fritz 9.0, 3.5 sec
playchess, 6.2 sec
and so on... What this records is the person switching between his chess program and the on-line connection that he/she is playing their game on. Interestingly, this recorded information is available to all computer busters at a popular chess site as well as the "time taken per move" and so on.
Of course RHP can't use this type of program for the 'move-a-day game', but don't ever think that other sites don't know what you are doing.
Originally posted by ArrakisAs an experiment, I renamed my rybka executable to be the same name as my anti-virus software, so it now appears in the Windows task manager with the same process name as my antivirus software. I also lowered the priority of this process.
you are mistaken to think that a determination can't be made if the person is using a chess engine while playing online at some sites
So how are you going to tell it's not my anti-virus that is running? Or are players with genuine busy non-chess software getting accused of cheating?
Originally posted by ArrakisI didn't mean that such an approach was not possible, but that it would be extremely inefficient and klunky. practically useless. it doesn't address the problem of recognizing engine moves in any way. in a relatively slow game, you could even use another computer for the analysing.
Well, I don't know about Playchess, but both of you are mistaken to think that a determination can't be made if the person is using a chess engine while playing online at some sites.
Originally posted by VarenkaTrust me- you can be caught. I will not divulge any more information because if I did it might help cheaters find a way around this scheme.
As an experiment, I renamed my rybka executable to be the same name as my anti-virus software, so it now appears in the Windows task manager with the same process name as my antivirus software. I also lowered the priority of this process.
So how are you going to tell it's not my anti-virus that is running? Or are players with genuine busy non-chess software getting accused of cheating?