Hey do you have the proof chess is draw? Have I missed it? :(
When Stockfish disagrees with a Grand master
Sort:
Seeing interesting unexpected results in nalimov tables, can't be sure about this one.
EvgeniyZh wrote:
Seeing interesting unexpected results in nalimov tables, can't be sure about this one.
man, if you loose an ending with 0.00 , y0u made a blunder in any moment.
Jul 30, 2015
0
#5
Trust the GM all the time until you are a master.
Stockfish is what you use when you don't have a GM in your house.
Jul 30, 2015
0
#6
excellent replies. 8) I have found the GM moves to be more sensible. Like in the game above I posted, stockfish, after 10 minutes and about 30 moves deep said that black should respond to 1. RB2 with 1. ... f4. I guess its important to remember the strategy in play more than 20+ move ahead tactics where many of the choices along the way result in similar strength positions. .15 point advantages are hardly crushing in human terms so a move taht maintains the same 'score' really is arguably just as good as a Stockfish move.
And I'll never be a master on this side of the grave. But its an absolute hoot trying to get there. I wish i could explain our collective fascination with it. Once I realized it was more powerful to play a strategy that helps choose what to calculate the game got a LOT more interesting.
Jul 30, 2015
0
#7
Which game are you guys looking at? Neither 1.RB2 nor 1...f4 are legal moves for the posted game sequence.
Jul 30, 2015
0
#8
yes rB1 and f5. 8/ I'm a mess and have been looking at that setup for a while. 5 of Swords -- why do you argue in favor of:
1. RB1 F5 ?
That's what Stockfish said to do, with I think 0.15 more 'score' than what the GM recommended which was 1. ... c5 to combat the fork attack of 2. RB7 against the c7 atd a7 pawns. Is it because of the importance you mention of the f pawn in rook endings? If so, I'll spend some time studying rook endings to try and get my head around that. 8) THANK YOU SO MUCH for replying everyone.
AWSmith61 wrote:
yes rB1 and f5. 8/ I'm a mess and have been looking at that setup for a while. 5 of Swords -- why do you argue in favor of:
1. RB1 F5 ?
That's what Stockfish said to do, with I think 0.15 more 'score' than what the GM recommended which was 1. ... c5 to combat the fork attack of 2. RB7 against the c7 atd a7 pawns. Is it because of the importance you mention of the f pawn in rook endings? If so, I'll spend some time studying rook endings to try and get my head around that. 8) THANK YOU SO MUCH for replying everyone.
What's usually good about GM analysis is that it almost always jives with general chess principles in a way that's typically not too hard to discern. Like in the shown example for White. Take the open file, threaten for an outside passer, control the 7th, mobilize the king, etc.
What's usually good with a computer view is that it's very efficient, generally spots well conceived counterplay, and loves tempo. As in the case ...f5 for Black which frees the Black King with access to center, quickly forces White to probably accept some weakness, and opens possible lines for the rook to try to equalize. Downside, often much harder to discover benefits.
Though in the position presented, I'm not sure Black's line ...c5, a5, Ra8 presented by the GM would actually be recommended or played by a GM. It appears way too passive and since likely losing anyway, why take the drawn out pain without trying for any counter?
KBBKN — Bernhard Horwitz and Josef Kling (1851) proposed that Black can draw by entering a defensive fortress, but tablebases demonstrated a general win, with maximum DTC = 66 or 67 and maximum DTM = 78.
KNNKP — Alexey Troitsky established this as a win for the knights if the pawn was blocked behind the Troitzky line. Analysis of the tablebases has clarified that even if the pawn has crossed the Troitzky line, White can sometimes win by forcing zugzwang. Maximum DTC = DTM = 115 moves.
BBPxNN has some interesting features. It appears as if it is usually a win if White has a knight pawn, but a draw for other pawns. For example, the position (Kb1,Bb2,Bc2,Pa2/Ka7,Nb7,Nc7) is drawn. If this position is shifted one file to the right, White wins, while if it is shifted 2 or 3 columns to the right it is a draw.
So you can never know. Maybe starting position is just one big zugzwang for White.
One huge advantage the GMs have over the engines is that they can actually explain things to you and I , the engines cannot explain anything . Its sorta like trying to learn math from a calculator ...
pfren wrote:
Sometimes it does not matter what tablebases say.
For example, the ending 2 bishops vs knight is now known to be a forced win, but in practice it will be drawn most of the times due to the 50 moves rule (assuming best play by both sides winning the knight or mating needs up to 66 moves).
That KBBKN scenario is really interesting. I'm wondering if there's actually any game examples where somebody tried to or suceeded in winning this?
So... lately, I've been looking at old Scotch games and running them through stockfish while studying them. I'm a ridiculous fan of the Scotch and enjoy wasting away reading about it. I even have 2 books dedicated to the opening. Whatever. I'm still merely 1050ish here, but I play for the sport, the adrenaline, and to occaisionally relish in a nice win that I can be proud of.
Now, that said, I am at a loss right now. I'm reviewing this:
http://www.chess.com/video/player/nominal-and-absolute-power-of-the-pieces
Which was written by GM Dejan Bojkov (a chess.com member whom I deeply respect). It takes guts to put your opinions out there for the world to critique and armed with a computer everyone's an armchair world chess champion.
I trust GM opinions MORE than those of Stockfish. Stockfish is complaining about the pawn structure here:
Anyway, how do you handle studying non-annoted games? I am running through a bunch of old scotch games, some of them very very old (1700s and 1800s) and they are BEAUTIFUL. But often Stockfish poo poos the moves in them. When it does, I find it tough to undrstand WHY.