Not quite. In the latest (Jan-03-15) CCRL 40/40 tournament list Stockfish 5 65-bit, 4 CPU is rated at 3284, 123 points higher than Rybka 4 64-bit, 4 CPU, rated at 3161. Perhaps more pertinent, in the latest (Jan-04-2015) CEGT 40/120 tournament list Stockfish is rated at 3094, 157 points higher than Rybka, rated at 2957.

Maybe I'm wrong, but if you start running chess engines for 24 hours, 48 hours or even 72 hours per move then I'm not sure we have a lot of data to make comparisons. Most chess engine comparisons that have been made are on short or tight time controls.

Rybka 4.1 at 72 hours per move would be a formidable opponent.>

No I think you are not wrong about the ability of Rybka and his human positions assessments.

But in a contest against us his success will largely depend on book data's then brute engines evaluations from R4.1.

He would need also a lot of computer time and personal effort for keeping up with the latest finds of theoretical novelties once he is out of book.

His playing repertoire may be of insufficient quality to compete against our unrevealed opening lines historical rated at super GM playing strength.

Thus if he keeps R4.1 he would have to out book us from the start and hope to stay within his prepared engine book lines until Rybka can take over.

Probably not nearly as much as you might think. Remember that the number of possible moves increases geometrically, and so does the time to advance one more ply. Everything depends on the capabilities of the hardware but if Rybka <in middlegame positions> can get to d=20 in 30 minutes, at the end of one hour it will probably only get to d=23 or so. And after 24 hours it would probably only get to d=28, after 48 hours it would probably only get to d=31 and at the end of 72 hours it would probably only get to d=33 or so. So the difference in search depth between 1 and 72 hours is probably only 5 plies or so.

These are just rough and likely pessimistic guesses on my part. <RandomVisitor> can of course come up with much more accurate numbers, particularly since he has occasionally run analyses for days.

Of course other engines would be laboring under similar constraints, but in <middlegame positions> it takes Rybka longer to reach a given search depth than it takes other engines. I say <middlegame> positions because in endgames, because of the reduced number of pieces, Rybka can get to depth in a similar amount of time as other engines, with the exception of Stockfish.

"Opening choice and preparation in OTB chess are important. The goal is to reach a known position that offers adequate chances, and, if lucky, the opponent goes astray somewhere. It also saves important reflection time. In CC opening choice is even more important. There are many openings that are suitable for OTB play, but are not strong enough in CC.

In CC it is very hard to win a game (our computer engine assistants are very strong). Hence, it is of the utmost importance to not lose a game. If you lose one game, you may have to play another five or ten games to achieve a win that compensates for the loss."

At the ongoing 28th World Championship Final, out of 108 finished games 98 ended in a draw..

https://www.iccf.com/event?id=37632

If not, that would make him (as far as I know) the first to brush the World Team off like that.

( I do not care if he mentioned it on his blog or whatever)

http://www.chess-server.net/uploads...

many of the participating players are fresh, ambitious and still undefeated (not only in that contest)

many .....>Just found this game there

[Event "BdF-Remote"]
[Site "ICCF"]
[Date "2013.2.10"]
[Round "-"]
[White "Mirbach, Thomas"]
[Black "Müller, Achim (Rheine)"]
[Result "1-0"]
[WhiteElo "2283"]
[BlackElo "2194"]
[Board "15"]
[WhiteTeam "BdF-Server"]
[BlackTeam "Remote"]

Sorry for taking so long to reply to you very valid question.

An answer is not easy to give, simply because your question is valid.

My attempted answer is this:

The final moves in any line are the easiest to explain, especially if the very last move is mate.

The hardest move to explain is the very first one, except where that move forces a reply, in which case it is the second move which becomes the hardest to explain.

:)

< < 12. BEST (WORST) PUN OF THE YEAR >

And the winner is ... <Suffering from c6-ness> >

Carolyn Wells, an author and poet who died in 1942, said,

"A blunder at the right moment

is better than cleverness at the wrong time."

From Phillip Adler's Bridge column today.

At least now I know why most of my questions are answered quickly! :-)

I guess that, given that several of us (or at least I) have a tendency to generate voluminous analyses (and some of us even find them useful), what I was hoping to find out was whether there was a way that the analyses could be presented in a form that would make them more easily comprehensible and approachable by those who have neither the time nor the inclination to review them in detail. You seem to indicate that you favor an explanation of the various moves in a line. But, in addition to taking more time and effort, if we were to attempt to explain or second-guess the computer about the moves, or even just selected moves, that would make the already voluminous analyses even more voluminous. And, of course, the quality of the explanations would be highly dependent on the skill and strength of the analyst. In my case any explanation would probably be or should be suspect until independently verified, and would require additional computer analysis of any suggested deviations from the original line.

Or, for example, would you prefer to see only the engine's top line listed and explained? I typically let the engine display its top 3 moves to see if there is one move that is clearly considered better by the engine than others and therefore the most likely to be played and the move where we should concentrate our efforts in analyzing further. Or whether there are 2 or 3 moves that are considered approximately equal and therefore we need to investigage them all. So perhaps I should be selective and in the first case only list the top line and in the second case list all the lines are evaluated approximately equal. That should reduce some of the clutter and "voluminosity". Yes, that's a word; I looked it up.