Only Chess
05 Feb 11
08 Feb 11
Originally posted by DiophantusThe OTB rating difference between these amateurs and their competitors was huge! Just how much do you think the amateur ratings were deflated by?
And yes, freestyle is different in some ways from CC but both share the “get the best out of an engine” (in cases where CC allows engines) aspect.
Apparently you have to be damn good at CC to get the best out of an engine in CC
One of the aspects of using engines for better analysis is to know the strengths and weaknesses of each. And then there’s the skill to get them to analyse in parallel while trying to refute each other. So if Rybka thinks Nc3 and Shredder thinks Bb4, you can start to play out lines against each other and see if the evaluations follow a trend. e.g. Shredder may start to reduce it’s initial evaluation of Bb4 as the line is stepped into. Of course, it may take many moves before a refutation is found, or it may be that both moves lead to the same game outcome. So it’s not straightforward. But someone who is good at performing such interactive analysis will be better than someone who just uses a single engine blindly.
I used to play at playchess.de which supports tournaments with and without engine assistance. My ranking with engine assistance was much higher than my ranking without, despite the fact I was playing many of the same players. So it wasn’t that our CC ability was changing but rather because of my interest and experience with computer chess. I could sometimes start to see which engine they were mainly using and how they were missing cases where another engine was producing a better move. I believe that the freestyle winners ZackS would thrash me in a similar CC game, simply because their ability to use the computer is so much better and little to do with our CC ability.
08 Feb 11
Originally posted by nimzo5How much better? They beat teams that included 2600+ GMs. They won the freestyle in 2005... you say he was rated something like 1950 in 2009... are we saying he may have been 2100+ in 2005? If not, there's over 500 rating points of a difference!
Cramton won the B prize at the world open the year after the freestyle win and was something like 1950 when he stopped playing otb in 2009. I used to play Blitz with him on icc, while not a master, he was much better than the rating chessbase had him pegged at.
08 Feb 11
1 edit
Originally posted by VarenkaI speak from my own experience, as do you. It seems our experiences differ which isn't a bad thing.
The OTB rating difference between these amateurs and their competitors was huge! Just how much do you think the amateur ratings were deflated by?
And yes, freestyle is different in some ways from CC but both share the “get the best out of an engine” (in cases where CC allows engines) aspect.
[b]Apparently you have to be damn good at CC to get the be ...[text shortened]... ecause their ability to use the computer is so much better and little to do with our CC ability.
I thought Playchess only offered blitz and standard time controls rather than CC style play? I couldn't find a way to play CC last time I went there although things may have changed in the past year.
08 Feb 11
Originally posted by DiophantusI was referring to www.playchess.de (not the Chessbase owned site), sorry for the confusion. And I accept our experiences differ.
I speak from my own experience, as do you. It seems our experiences differ which isn't a bad thing.
I thought Playchess only offered blitz and standard time controls rather than CC style play? I couldn't find a way to play CC last time I went there although things may have changed in the past year.
08 Feb 11
Originally posted by VarenkaMaybe not for me then. My rating where computers are allowed hovers around 1700, bottom end of the middle orders who use computers but ineffectively and top end of the computer illiterate dinosaurs at the bottom of the food chain. You'd think that a chap who has access to a couple of super computers in his day job could do better than that.
May be worth a look if you're interested in computer assisted games (their ACL league) but I'm not sure how their membership is doing these days.
08 Feb 11
Originally posted by VarenkaI am simply stating that in 2005 he was stronger than 1650 or possibly the prep for the freestyle made him stronger.
How much better? They beat teams that included 2600+ GMs. They won the freestyle in 2005... you say he was rated something like 1950 in 2009... are we saying he may have been 2100+ in 2005? If not, there's over 500 rating points of a difference!
Freestyle has it's own idiosyncracies and quite a bit of what makes a GM a strong player may not translate into any distinct advantage in this format. It's far more important to have a well developed opening book and have team members who are extremely good at doing rapid analysis. It wouldn't surprise me if most of the strongest teams now are one decent player and a bunch of comp chess guys backing the player up.
08 Feb 11
Originally posted by DiophantusThis is an illusion, it's the justification used by those who rely on engines in their CC matches. I don't say that good CC results can be achieved without work and skills, but those skills are not related to chess. Almost all the battle takes place in the opening: you must use a very good database and you must explore further some lines to overcome the horizon effect of your engine.
Freestyle isn't quite the same thing and I wouldn't think the ZackS team's USCF ratings are a true reflection of their freestyle abilities. If they spend a lot of time playing freestyle they are not likely to be entering many standard tournaments to get their ratings up. The freestylers I know generally ignore standard OTB chess because they find the man- ...[text shortened]... id to "Apparently you have to be damn good at CC to get the best out of an engine in CC".
08 Feb 11
Originally posted by cotoiSo you think freestyle is the same as CC? I don't see how that could be used as a justification for using an engine in CC matches.
This is an illusion, it's the justification used by those who rely on engines in their CC matches. I don't say that good CC results can be achieved without work and skills, but those skills are not related to chess. Almost all the battle takes place in the opening: you must use a very good database and you must explore further some lines to overcome the horizon effect of your engine.
I agree that modern CC is more about the opening than other phases of the game, although I have seen one or two who rely overly much on the engine get clobbered in the endgame. This seems to carry over into the less engine intensive areas as well now.
08 Feb 11
Originally posted by cotoiDo you mean ICCF style cc?
This is an illusion, it's the justification used by those who rely on engines in their CC matches. I don't say that good CC results can be achieved without work and skills, but those skills are not related to chess. Almost all the battle takes place in the opening: you must use a very good database and you must explore further some lines to overcome the horizon effect of your engine.
08 Feb 11
Originally posted by nimzo5It's not the horizon effect, because this would only refer to the length, it's much more and nearly exclusively evaluation and interpretation of a given position. No engine would actually include in their calculation the sacrifice of a piece which will only lead to a long lasting weakness on the dark squares but which would be decisive for the outcome. Every experienced and highly skilled player would, because he knows about the importance of weak squares. But on a high - a very high - level i.e. the actual World CC Championships (including the Candidates Finals all played on the ICCF server) 90% of the results are draws. Guess why!
Do you mean ICCF style cc?
09 Feb 11
Originally posted by DiophantusIn freestyle strong OTB players are more likely to get an advantage over patzers, due to time constraints. The link posted above shows that engine are so powerful that they can nullify the experience of a strong OTB player.
So you think freestyle is the same as CC? I don't see how that could be used as a justification for using an engine in CC matches.
.