Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / Grandmasters play cautious
- - By apollox (***) [gb] Date 2014-01-23 03:04
In studying the games of the recent Anand - Carlsen match, I noticed that their choices tended to be less forceful than the top selections of the chess engines. Their moves always avoid the lines that Rybka, Houdini, Critter, etc. would invariably go for. In most cases, the high level engines, even the free-downloadable ones, see almost identical best lines. However, the best lines seem inflexible. If a human were to follow these sharp lines, the steps have to be correct with little room for deviation. I can see now how these GMs think. Keep your options open, with a draw always within grasp.
Parent - - By M ANSARI (*****) [kw] Date 2014-01-26 07:10
Carlsen has probably the best explanation I have yet to hear about openings ... that is that generally an opening has a lot of margin of error.  At some point that margin of error is reduced or you end up with a possibility that one line is forced due to some obscure tactical reality (although not visible to a human).  When analyzing with a strong engine with super strong hardware, it is just incredibly amazing what resources an engine can find in a position ... just incredible.  Chess is a lot richer that most people realize ... humans are like an entity  navigating through a beautiful scenery with +20 vision, we can appreciate the beauty from far but can be killed by a low branch or a well hole that is only a few inches away that we couldn't see.  Sure we might stumble through all the hazards successfully  sometimes, but we will never appreciate all the dangers that we would have faced that were along the path.
Parent - - By apollox (***) [gb] Date 2014-01-26 19:22

> Chess is a lot richer that most people realize

I am astounded at the sheer number of out-of-date annotations in my old chess books when I test the engines on them. For example the antiquated annotation giving a "?" to move followed up with "better is... etc." turns out to be erroneous on numerous occasions. Golombek's analyses in Capablanca's Hundred Best Games of Chess is riddled with what we know now are dubious comments because of the engines of today. In club chess, it probably makes little difference and these old chess books still retain value for seeing the tactics and strategies of the old masters. I would still have an engine as a companion with these books however.
Parent - By Mark (****) [us] Date 2014-01-26 22:55
I used to enjoy going through my old endgame books and grandmaster games when I first got the 6-man tablebases. There's no end to the number of major errors you can find!
Parent - - By Uri Blass (*****) [il] Date 2014-06-05 10:29
I think that a book with analysis by some GM without engines when the GM does not give only
lines and evaluation but also the level of confidence that he has in his opinion can be good.

I would like to read comments like I am 90% sure that move A is better than move B when hopefully analysis later shows that in 9 out of 10 cases the GM is right.

Bigger confidence can be 99% or even 100% and of course the GM can get 100% confidence if there is some simple tactical line.
Parent - By Carl Bicknell (*****) [gb] Date 2014-06-05 10:53
Books are meant for people though.

From an objective sense what you're saying is true, but over the board chess has a lot to do with pressure, and creating favourable practical chances. As yet the machines can't comment on this.
Up Topic Rybka Support & Discussion / Rybka Discussion / Grandmasters play cautious

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill