It seems that both accusers and defenders of Ih8sens are telling us to match up his moves against an engine. As if it were that simple. I recently acquired Fritz 10. A couple of days ago I decided to look at an opening variation I'm interested in both with Fritz 10 and an earlier version I had (5.32). I couldn't believe how differently they seemed to analyse and how often they came up with different moves. (My initial impression is that Fritz 10 places more value on initiative and attack, Fritz 5.32 on material.) And recently, a strong player told me that Rybka (which I've never used) very often comes up with different moves from Fritz, apparently having a better feel for positional play and strategy (definitely one of Fritz's weak points). And of course, the move you will get from any engine will depend on how long you let it look at the position (plus other factors such as computer processing speed, RAM, hash tables set etc).
I think the mods have already told us that no one has ever been banned on the basis of one move or even one game, so the game that Scandium showed us is not relevant. (For what it's worth - I'm an experienced KG player myself - the game seemed perfectly 'human' to me.) I should imagine there are many games where the loser may have a higher engine match. Quite often when your opponent hits you with a strong combination and starts walking all over you, you have a depressing series of virtually 'only' moves, whereas your opponent may have a number of different ways to win.
I don't wish to make any other comment on the guilt or otherwise of ih8sens (except the basic assumption of innocence till proved guilty), since I don't have the evidence available to make an informed judgment. I suspect that few of us contributing to this thread do either.
Originally posted by Dragon FireI didn't want to get back into this thread, but this seems to me pretty absurd. You're saying that if someone matches up with an engine even for just three consecutive moves after leaving book, he is likely to be cheating? I think this is the easiest time to match engine moves because the main ideas from the book still apply so your moves will likely be very natural. In fact, you may play better than an engine. Furthermore, it really matters what kinds of positions they are. Sharp positions are more forced and so of course match-up may be higher. And of course, the strength of the player should be taken into account.
Believe it or not I think we agree more than we disagree.
I think we both know that if someone matches up with engine moves, even only 2-3 moves deep at the point where they leave known books and DBs in virtually all their games they are not using engine research but are using an engine.
It is precisely at this point that most players are most like ...[text shortened]... significant matches in a majority of games then we are indeed worlds apart and will never agree.
I mean, if the standards are so low, it would be assumed that I cheated in many of my OTB games. Certainly, they tend to be better than my RHP games because I spend much more time on my moves and sometimes my match-up is quite high with Rybka. Take the two hour game I just played THIS evening. It is definitely not among my best games and I left opening book by move four when he played Bb4. However, only 3 moves are more than .10 away from Rybka's first move choice with 20 minutes per move and even those are very close and could well come down to preference (e3 vs b4 and Rc3 vs Kc5 and Rf3 vs e6) and occurred mostly in time trouble. Here's the game anyway (I was White against a 1700):
1. c4 f5 2. g3 Nf6 3. Bg2 e6 4. Nc3 Bb4 5. Qb3 c5 6. a3 Bxc3 7. Qxc3 Nc6 8. d3
Qa5 9. Bd2 Qxc3 10. Bxc3 O-O 11. e3 a6 12. b4 b6 13. b5 axb5 14. cxb5 d5 15.
bxc6 Ba6 16. Ne2 Rac8 17. Nf4 Rxc6 18. Bxf6 Rxf6 19. Nxd5 exd5 20. Bxd5+ Kf8
21. Bxc6 Rxc6 22. O-O-O Rd6 23. d4 c4 24. Kc2 Bb7 25. Rhe1 Bf3 26. Rc1 b5 27.
Kc3 Ra6 28. Kb4 Bc6 29. Rc3 Ke7 30. d5 Bd7 31. Kc5 Ra8 32. f3 Rc8+ 33. Kd4 Kd6
34. e4 fxe4 35. fxe4 Re8 36. e5+ Kc7 37. Rf3 Ra8 38. Rf7 Kd8 (the rest is not interesting and the game is clearly over) 1-0
So tell me, would this game be suspicious on RHP? Would I be banned after several games like this? I hope the mods have higher standards than you.
Originally posted by exigentskyNot just a single game but consistently matching up with an engine when you leave "book" is, in my opinion, highly indicitive (not in itself proof) of engine use as it is precisely at this point that mistakes are most often made.
I didn't want to get back into this thread, but this seems to me pretty absurd. You're saying that if someone matches up with an engine even for just three consecutive moves after leaving book, he is likely to be cheating? I think this is the easiest time to match engine moves because the main ideas from the book still apply so your moves will likely be v d I be banned after several games like this? I hope the mods have higher standards than you.
Indeed it is at this point that ih8sens made an error in the Kings Gambit we played back when he was rated 1600.
Originally posted by tomtom232I appreciate that, but yet that's what makes this all the more important.
I also think that kmac27 would attest to your innocence witty.
As has been said by others, the RHP community is helpless to change a decision to implement a ban (although I find it somewhat ironic that it often has complete power in initiating the scrutiny/allegations).
Originally posted by searchingforkolischNot really any way to tell, because Playchess has this crappy thing that won't allow you to view a players past games...
Just curious. Not referring to his ratings but does his playchess games resemble the ones he played during the massive rating jump?
We can only judge by ratings, and doing that you'd have to assume that he plays chess at the same level as me - my peak on here was 1725, and I got seriously outplayed by opposition over 1700.
Originally posted by cmsMasterFrom:grater216
Not really any way to tell, because Playchess has this crappy thing that won't allow you to view a players past games...
We can only judge by ratings, and doing that you'd have to assume that he plays chess at the same level as me - my peak on here was 1725, and I got seriously outplayed by opposition over 1700.
Date:Jan 18 2008 10:17
Subject:Someone asked about my playchess games..
Here's the last playchess game I played.. it was 3/0 so I obviously have a dual core computer so as to run my engine as well at playchess at the same time ... plus I'm a really fast clicker
anyways.. here it is.. I've analyzed it slightly and it looks good.. I'm not sure if it will help my case or not though.
[Event "Rated game, 3m + 0s"]
[Site "Main Playing Hall"]
[Date "2008.01.17"]
[Round "?"]
[White "Brainless00"]
[Black "Ih8sens"]
[Result "0-1"]
[WhiteElo "1444"]
[BlackElo "1489"]
[PlyCount "52"]
[EventDate "2008.01.17"]
[TimeControl "180"]
1. e4 {1} e5 {1} 2. Nf3 {2} Nc6 {0} 3. d4 {1} exd4 {2} 4. Bc4 {2} Nf6 {2} 5.
Ng5 {3} d5 {14} 6. exd5 {1} Na5 {5} 7. Bb5+ {1} c6 {1} 8. dxc6 {1} bxc6 {1} 9.
Bd3 {1} h6 {1} 10. Nf3 {4} c5 {7} 11. O-O {2} Be7 {3} 12. c3 {7} c4 {2} 13. Be4
{9} Nxe4 {3} 14. Re1 {5} Nf6 {10} 15. Nxd4 {2} O-O {3} 16. Na3 {12} Bb7 {2} 17.
Qc2 {20} Qd5 {4} 18. Nf3 {8} Bxa3 {12} 19. bxa3 {5} Rfe8 {5} 20. Bf4 {11} Nh5 {
18} 21. Be5 {4} Nc6 {14} 22. Bd4 {8} Nxd4 {3} 23. cxd4 {3} Nf4 {6} 24. Rad1 {0}
Nxg2 {2} 25. Ne5 {29} Nxe1 {6} 26. Rxe1 {0} Qg2# {(Lag: Av=0.71s, max=1.2s) 2}
0-1
If you wanna post that in the forum for me I'd really appreciate it but I certainly won't mind if you want to stay out of this .
Originally posted by cmsMasterNo offense, but while possible, there are very few people that good at cc, and that bad at blitz
From:grater216
Date:Jan 18 2008 10:17
Subject:Someone asked about my playchess games..
Here's the last playchess game I played.. it was 3/0 so I obviously have a dual core computer so as to run my engine as well at playchess at the same time ... plus I'm a really fast clicker
anyways.. here it is.. I've analyzed it slightly and it looks good. ...[text shortened]... te it but I certainly won't mind if you want to stay out of this .
For instance I'm anywhere from a 1550-1700 on here and on playchess my blitz fluxuates from 1550-1750 (some days I just don't see anything).
Originally posted by zebanoWell, that's true, but I don't think we should read too much into CC and blitz rating correlation. I'm probably between 1650 and 1750 RHP and between 1400 and 1500 blitz.
No offense, but while possible, there are very few people that good at cc, and that bad at blitz
For instance I'm anywhere from a 1550-1700 on here and on playchess my blitz fluxuates from 1550-1750 (some days I just don't see anything).
Slow game ratings may have a correlation though.
This has happened (at least once in the olympics, several times, at least, at Boston, and probably innumerable times all over the world):
marathon race (26.2 miles). world champions finish in just over 2 hours (2:08 - 2:12); world class runners finish in under 2:30.
Situation: an overweight, out-of-shape runner (by this we might mean only 10 Lbs/5 Kg) closes in on the finish line to come close to winning, or even winning, the race. the 1st prize award is NEVER granted to the imposter.
Rationale: the runner "couldn't have done it." impossible. of course, the race typically has video records at multiple check points - but even those could miss a runner, and it would take weeks to "prove" that a runner hadn't passed them, and a runner theoretically - not really - could have snuck into a number of the video sequences in appropriate positions all along the race. the runner NEVER wins that argument.
In the real world, sometimes validation after the fact is required. For instance, you get a perfect score on the math SAT, they might have you sit for a monitored validation exam, only to find out that you can't do simple algebra, let alone trigonometry and advanced calculus. conclusion: you cheated.
The key is consistency. it is possible to improve to a master strength real fast. yes, there are sub-10 year old masters and sub 15 year old grandmasters in real life. however, no master falls back to 1200 level when they are confronted in a real life "monitored" game.
conclusion: how can RHP tell rapid improvement from engine cheaters? (my response, not an official RHP answer): engineers that would consistently use engines, and ideally different ones to make different moves all the time, could not be caught. but then, i don't think that they are the problem. they are just a engine using idiots pretending to be human. i could care less. however, an engine using idiot that pulls out the engine to win tournaments and clan games, then goes back to human moves (1100-1500) for those regular old individual or unimportant games, exposes their engine use and cheating. the only issue is how to catch them. well, they get reported as suspicious. the admin/mods take a look at the pattern of their games and see that sometimes they are a 1400 player and sometimes they are a master. ergo, two different players, ergo, cheater.
the question posed in this thread that interests me is: do the interests of the greater RHP community come into play? do the rules of "due process" apply? in the USA, someone accused has a right to confront the accuser and the evidence. no one in RHP openly knows how RHP handles this. i'm sure that some do, as ih8sens may be with us today using some other name. i doubt that the player would come forward and tell us in an open forum.
one final point. I joined a little over a year ago. there was a curious thread at that time about a user that had been banned for multiple accounts and who was allowed back in. so, there must be some appeal and dialogue process that is in place. if you have the chance to defend yourself and make a case for how you aren't an engineer, then those concerns might be answered. should this be a public process? i don't know.
Originally posted by coquettethey don't. when a user joins rhp, he agrees the site has the right to terminate his account if they feel like it.
do the rules of "due process" apply?
13. TERMINATION
You agree that RHP may, under certain circumstances and without prior notice, immediately terminate your RHP account,and access to the Service. Cause for such termination shall include, but not be limited to, (a) breaches or violations of the TOS or other incorporated agreements or guidelines, (b) requests by law enforcement or other government agencies, (c) a request by you (self-initiated account deletions), (d) discontinuance or material modification to the Service (or any part thereof), (e) unexpected technical or security issues or problems, and (f) extended periods of inactivity. Termination of your RHP account includes (a) removal of access to all offerings within the Service, (b) deletion of your password and all related information and content associated with or inside your account (or any part thereof), and (c) barring further use of the Service. Further, you agree that all terminations for cause shall be made in RHP's sole discretion and that RHP shall not be liable to you or any third-party for any termination of your account, any associated email address, or access to the Service.
Originally posted by coquetteYour rational is valid within reason.
The key is consistency .... how can RHP tell rapid improvement from engine cheaters? .... engineers that would consistently use engines, and ideally different ones to make different moves all the time, could not be caught .... pulls out the engine to win tournaments and clan games, then goes back to human moves (1100-1500) for those regular old individual ...[text shortened]... y are a 1400 player and sometimes they are a master. ergo, two different players, ergo, cheater.
Firstly anyone who plays at 1100 strength one day and 2000 the next is almost certainly a cheater because nothing will make a 2000 player play that bad (not even a night drinking, except the odd move), but if someone plays 1700-1800 in one set of games and 2000-2100 in another that is not necesarily the case as in more important games they may use databases, books and simply take their time whereas in less serious games they throw these out the window just to see how they do, they may even be blitzing their games or confident of a win just having a bit of fun. Same person, different style, different strength!
Originally posted by coquetteNo. RHP is a private company and they can do whatever they want (within the respected laws that affect them). They could let the public vote and make decisions for them I guess but...
...should this be a public process? i don't know.
When it comes to judging who is cheating and who is not, are we going to let a random person(s) on the internet who may not really know what they are doing make a decision or at least have impact in a decision that has real life consequences or should a person who moderates this site and has had experience with detecting real cheating figure out who is cheating?