Only Chess
06 Mar 08
10 Mar 08
Originally posted by no1marauderNo. I said that correspondence comparisons should be made with correspondence masters. Wormwood quickly responded above and said that this is indeed what RHP has done.
So you think players on a correspondence site routinely play better than Kasparov did OTB?
10 Mar 08
1 edit
Originally posted by NordlysI totally understand where you and wormwood are coming from on this. I was really interested in how much higher the match-up rates are now compared to the pre-computer era cc masters. When is the pre-computer era for cc? Is it considered to be before 2000 or so? I really would like to see what someone in the RHP testing got for match-up rates for recent Berliner games versus whatever pre-computer cc master that is a counterexample being used.
I have no idea what the match-up rates are, but I think what wormwood meant was that whatever the match-up rates of modern CC masters are, it won't help us as a guideline because the ICCF allows engine use.
EDIT: if in order to answer these questions the transparency of the testing procedure might be compromised, then don't answer. maybe give me a ball park figure in private message?
10 Mar 08
Originally posted by Doctor RatAnd I say you're wrong.
No. I said that correspondence comparisons should be made with correspondence masters. Wormwood quickly responded above and said that this is indeed what RHP has done.
First, for the rather obvious reasons already given i.e. that modern correspondence players use engines and have done so for a very long time.
Second, that although this is a correspondence site, players here don't play like correspondence players did in the 70's and 80's. Those players played in very few games and often took years to complete them. Neither of those conditions generally apply to the players on this site.
If you want to research correspondence games for 40 years ago and run them through Fritz, go ahead. I can pretty much guarantee that they won't routinely reach 75-80% matchups. If they did, Berliner and the rest wouldn't bother to use engines, would they?
10 Mar 08
I think that people who play with chess engines, should own up and have a site of their own so they can all "PLAY" together.
Alternatively, if they play with an engine, then people have to beat the engines! otherwise mankind will not survive, er.... or something....
10 Mar 08
1 edit
Originally posted by no1marauderWe found that the match-up rates of pre-computer era CC GMs and modem OTB GMs are very similar. Even super GMs are only about a percentage point or so higher than the norm. All strong verifiably human play exists in quite a tight band.
And I say you're wrong.
First, for the rather obvious reasons already given i.e. that modern correspondence players use engines and have done so for a very long time.
Second, that although this is a correspondence site, players here don't play like correspondence players did in the 70's and 80's. Those players played in ver % matchups. If they did, Berliner and the rest wouldn't bother to use engines, would they?
For matching just the top move, over many games, you are looking at a control stat of 50%-60% depending on the nature of the games. The most blatant offender we've ever had (Jean Hebert, the imposter, remember him?) managed over 75% on first choice, over many games, some 20% above the control set!
We've never considered running tests on modern day CC players because we could never preclude the possibility that they were computer-assisted. They don't provide useful control data.
But I do admit it would be an interesting exercise to see if we'd ban them here.
11 Mar 08
Originally posted by no1marauderPlease site some sources that give examples of correspondence chess champions using computer assistance for their games. I'm sure you understand that a lot of hand-waving shouldn't be taken as fact by anybody, so I'm just looking for the same sources of info that you have that gives you the confidence to say that with such certainty. And please, none of that everybody just knows that to be true type of argument. I am seriously looking for articles/websites that talk about computer use in modern day games of correspondence champions.
And I say you're wrong.
First, for the rather obvious reasons already given i.e. that modern correspondence players use engines and have done so for a very long time.
Second, that although this is a correspondence site, players here don't play like correspondence players did in the 70's and 80's. Those players played in ver ...[text shortened]... % matchups. If they did, Berliner and the rest wouldn't bother to use engines, would they?
Concerning checking out past champions against software %'s, I think that sounds like an interesting project, and I was under the impression that RHP already did that, as per wormwoods post above. No1m, are you saying that RHP has, or hasn't, corroborated or tested software %'s for any cc masters past or present?
11 Mar 08
1 edit
Originally posted by GatecrasherThat is very interesting stuff! Thank you for your post, Gatecrasher.
We found that the match-up rates of pre-computer era CC GMs and modem OTB GMs are very similar. Even super GMs are only about a percentage point or so higher than the norm. All strong verifiably human play exists in quite a tight band.
For matching just the top move, over many games, you are looking at a control stat of 50%-60% depending on the natur l data.
But I do admit it would be an interesting exercise to see if we'd ban them here.
EDIT: I want to match up some Oosterom games and see what shakes out 🙂
11 Mar 08
Originally posted by Doctor RatMy Chess Life (published by the United States Federation) that had an interview of Berliner had him admitting to it. It's simply not hand waving. Google if you want more information.
Please site some sources that give examples of correspondence chess champions using computer assistance for their games. I'm sure you understand that a lot of hand-waving shouldn't be taken as fact by anybody, so I'm just looking for the same sources of info that you have that gives you the confidence to say that with such certainty. And please, none of t ...[text shortened]... HP has, or hasn't, corroborated or tested software %'s for any cc masters past or present?
11 Mar 08
1 edit
Originally posted by zebanoI will try to read more about that. I tried googling it before, but came up short, but with more terms (thank you for your source!) I will be more successful. Thank you for your quick response.
My Chess Life (published by the United States Federation) that had an interview of Berliner had him admitting to it. It's simply not hand waving. Google if you want more information.
EDIT: I have googled for Hans Berliner admitting to using computer assistance during a game. I cannot find anything. I have found his quote, "I have sort of independently come to the conclusion that cc is about played out. Anyone who loses a game with White clearly does not know much about chess, how to use databases, or does not have a state of the art computer..." (Hans Berliner) This is not an admittance of using computer assistance during a game. Berliner is talking about using a computer as a tool between games.
Until someone can show me a source where a CC Champion has admitted to using computer assistance during a game, or a source where someone has proven that computer assistance was used, I will maintain that your HAND WAVING does not convince me. (I'm not being a jerk; I am simply trying to be reasonable in requesting a hard source for such an accusation of cheating, I'm sure you understand)
11 Mar 08
1 edit
What would also be interesting is comparing matchups of strong CC players as they played throughout the modern era. Players like Berliner were champions in the 50's and 60's, well before a computer would be of any real use. If all CC GMs started using computers (I doubt this is true in all cases), we ought see a general rise in matchup percentage.
** EDIT ** I forgot I was posting in this thread. I hope it goes away soon. 😛
11 Mar 08
Originally posted by DawgHausI guess the trick is to find enough games from the players to study!
What would also be interesting is comparing matchups of strong CC players as they played throughout the modern era. Players like Berliner were champions in the 50's and 60's, well before a computer would be of any real use. If all CC GMs started using computers (I doubt this is true in all cases), we ought see a general rise in matchup percentage.
** EDIT ** I forgot I was posting in this thread. I hope it goes away soon. 😛
11 Mar 08
1 edit
Originally posted by Doctor Rattry harder, it took me less than 15 seconds.
Until someone can show me a source where a CC Champion has admitted to using computer assistance during a game, or a source where someone has proven that computer assistance was used, I will maintain that your HAND WAVING does not convince me. (I'm not being a jerk; I am simply trying to be reasonable in requesting a hard source for such an accusation of cheating, I'm sure you understand)
http://amici.iccf.com/issues/Issue_07/issue_07_ivar_bern_part_1.html
11 Mar 08
1 edit
Originally posted by Doctor RatI think it is. You don't need a state of the art computer just to cull through a database. You do need one if you're going to beat the engines the others players are using.
EDIT: I have googled for Hans Berliner admitting to using computer assistance during a game. I cannot find anything. I have found his quote, "I have sort of independently come to the conclusion that cc is about played out. Anyone who loses a game with White clearly does not know much about chess, how to use databases, or does not have a state of the art c ...[text shortened]... ter..." (Hans Berliner) This is not an admittance of using computer assistance during a game.
Edit: I mean, what's the point of even talking about 'admitting' computer use when ICCF allows it?! Yeah, next I'll "admit" to flagging people over-the-board in dead lost positions.