Reti's Study
Sort:
Aug 12, 2007
0
#2
i feel so inferior when i see things like this. when i look at this puzzle it never even crossed my mind that the black king is somewhat hemmed in. amazing.
amazing puzzle
Very nice and instructive. Consider it adopted for my file as well. Thanks Bill and GM Reti.
underpromotion!
4....e1(promotes to knight) stops the mate threat and black lasts at least a few more moves. May even has some stalemate possibilities. Don't have time to go into move sequences, analysis. I am busy preparing for a tournament. Just taking a break and happened to notice the flaw in the study.
I think if 3. .... e1/N then 4 Rf2+ Ke3 5 Re2+ is a clear win.
Phil_from_Blayney wrote: I think if 3. .... e1/N then 4 Rf2+ Ke3 5 Re2+ is a clear win.
My point was not whether it is a forced win for White. Rather an endgame study should have a decisive conclusion within 5 to 10 moves. Reti's composition is more like an endgame problem.
I might be wrong, but I don't think Reti took the underpromotion to a knight into account. More than likely someone else prior to me noticed this error.
Your move sequence, with optimal White and Black moves, will lead to a win for white, but only after 14 moves. I also checked it on Namilov,Rybka, Shredder, and Crafty Endgame Tablebases. All give 14 moves for White to win after 4.Re2+ 5.Kf3. All of the tablebases also give some drawing lines. That is another 1/2 of a normal game. And, if sub-optimal moves are made by White, it just increases the number moves needed to win and the possibilities of drawing.
tonydal,
A question for you: When you looked at the position, did you see the possiblity of the underpromotion to a Knight?
I'm pretty sure you're wrong, transpositions. A guy like Reti (a world class GM as well as one of the greatest endgame composers) would NEVER overlook an underpromotion--indeed, those guys live for stuff like that! An endgame study=forced win for White. There is no difference between an endgame study and an endgame problem.
I would agree if you were right. There is a diference between and endgame study and endgame problem. An endgame study is composed. An endgame problem is from an actual game.
Flaws have been found in many endgame positions (i.e. Rueben Fine's Basic Chess Endings--Ask GM Larry Evans) as well as middle game and opening positions.
There is one thing that I have learned in life and that is never say NEVER. Also if it matters that much to you then try posting some analysis and/or checking the tablebases.
My point is think for yourself, and then complement or constructively criticize. Don't post "in awe, gaga faced" approving comments. It helps no one and insults everyone's intelligence.
It is nice to know that you share the preference of reaching a conclusion in 5 to 10 moves.
You are right there is no law against NOT thinking for yourself. Everyone is entitled to do as they please as long as it is not illegal. I would add immoral, but that category has become grayer and grayer for all of us.
I have tried to find Reti's study online to see if he gives any analysis regarding the underpromotion. If Reti didn't, that fact would begin to narrow the specualtion. Haven't been able to find it. This reminds me of Fermat's Last Theorem, where he left only a sketchy note on the margin of a page claiming to know the proof. Even today some professional mathematicians still believe that Fermat's solution was much shorter and simpler than Andrew Wiles's. solution.
nimzovich wrote:
Apparently Arthur Mandler, an endgame analyst and close friend of Reti, published a collection of Reti's works two years after the latter's unfortunate demise in 1929. I wonder if Mandler offers any insight on Reti's intentions for his compositions, but this published work may be hard to obtain and beyond reasonable attempts at research.
The internet can sure be hit-or-miss in terms of research.
The mysterious and romantic stories Fermat's Last theorem is certainly another kettle of fish! I heard a less popular theory, with minimal following, speculating that Fermat made an error in his writing due to dyslexia!
Very nice piece of research, thank you. How did you come by this information?
I found the book in one location. It is in the J.F.K. Memorial Library at Cal. State University at Los Angeles in the Munitz Chess Collection. That is assuming it hasn't been stolen or lost. I will let you know if I get my hands on a copy.
A funny story. My very first chess book, My System, by Aaron Nimzowitsch, given to me by a fellow chess player, was a mustard yellow hard back copy that had been stolen from the Dallas Public Library. I mailed it back to them of course, and immediately purchased a copy.
Thanks again
tonydal wrote: I too am not interested in starting up some war over this point. Never say never is generally my policy too...but saying that a guy like Reti (not god, by the way--guy) could overlook an underpromotion in one of his own compositions is about like saying that he could fall into the scholar's mate. I am as irreverent and unlikely to be awestruck as they come--but there ain't enough whisky in the world to get an endgame composer like Reti so drunk that he'd overlook such a move.
No, but maybe his regular drinkin' buddy, Alekhine, could have. Just joking.
Read the posts between nimzovich and myself. You will see I am looking into the matter. I will keep you posted.
I had never considered an endgame problem to require more than a clear winning position, except when it finished in a mate and may even be a mate in two or three etc problem.
In this case, the given answer led to a lengthy rook endgame win. So the Knight underpromotion couldn't be considered any better because it leads to a win in a comparable number of moves, maybe even less. That it wasn't mentioned in the solution may just be an oversight, but until someone finds the problem and answer given by Reti, I shall presume that it has been an error by a later human presenting the problem.
nimzovich,
I just googled with: "Arthur Mandler" see for yourself:
http://opac.calstatela.edu/search/?searchtype=c&searcharg=GV1445+.R4+1948
Just a thought, maybe Bill Wall knows the answer. He was after all the one who started all this. Whoever knows him best maybe could send him a message.
My collection consists of at least 100 books, but my prized possessions are:
My System (THE HOLY GRAIL!!)
Die Blockade
Chess Praxis
Pawn Power In Chess, by Hans Kmoch
Yuri Averbak's Engame Series(5 volume set) and the recently published CD
Basic Chess Endings, by Rueben Fine (edited with all of the corrections dating from the 1970's in Chess Life and later Chess Life & Review in GM Larry Evans column)
I will keep you posted
Thanks again
In this case, the given answer led to a lengthy rook endgame win. So the Knight underpromotion couldn't be considered any better because it leads to a win in a comparable number of moves, maybe even less. That it wasn't mentioned in the solution may just be an oversight, but until someone finds the problem and answer given by Reti, I shall presume that it has been an error by a later human presenting the problem.
For the text highlighted in red, could you please cite the publication that you are referring to.
We are looking into the matter and your help would be appreciated.
Thanks
To transpositions - you may need to read my post again, all of it was just my humble opinion. 
I will look through the few endgame books that I have to see if I have any further information that may help. I can't remember having seen it before, but that is definitely no guarantee that I haven't. 
Thank you
transpositions wrote:
This reminds me of Fermat's Last Theorem, where he left only a sketchy note on the margin of a page claiming to know the proof. Even today some professional mathematicians still believe that Fermat's solution was much shorter and simpler than Andrew Wiles's. solution.
It has been a while since I read "Fermat's Enigma" by Simon Singh, but I recall that he put forward a nice possibility of what Fermat's "solution" might have been. Obviously it was not Wiles's solution, which depends on very obscure modern mathematical constructions. But apparently not long after Fermat's death, someone published a solution that seemed to be correct, until someone else pointed out that it only worked if (as I recall) imaginary numbers were excluded. I think Singh's best explanation for Fermat's assertion was that Fermat had found this (almost correct) proof and had not noticed that it did not work if one included imaginary numbers.
In Chess Life recently Pal Benko (as I recall) published some of Reti's endgame studies that had flawed solutions, and Benko modified them to be "correct" within the spirit of Reti's study.
Giants can certainly overlook things.
Here is a study by Richard Reti.