I can't solve this Retrograde analysis. Can you help?
Alan Trzuskoski
I have been assigned to solve this retrograde analysis problem for
class. Only i am just a beginner at chess and cant seem to solve it.
There are 96 moves that precede this one.
Any help that you could give would be great.
George suggests the following as the first fewmoves backwords:
The last move had to be the white queen blocking check from black rook.
If you look at possibilities, you see that no other move was possible. I
assume white moved last. But since there is only one possibility, and
that theory works.White's rooks can't have moved, nor could the bishops
and knight on a1. The knight on c7 couldn't have moved there since
black's king would be in check if the knight wasn't there. The only
other possibility is the queen moving to block check from b6. I have no
idea what the next moves are.
Thanks to all for helping
Alan
Duncan Harwood
"Alan Trzuskoski" <trz...@rpi.edu> wrote in message
news:3BDEE546...@rpi.edu...
What is your playing strength?
Alan Trzuskoski
mik...@unsafe.nu
to have been a piece on a7 that the queen captured. Otherwise, how
would the rook have gotten to a6?
On Tue, 30 Oct 2001 17:38:38 GMT, Alan Trzuskoski <trz...@rpi.edu>
wrote:
Duncan Harwood
board (white king is on a8 not h1). Also you never said whose turn to move
it is.
Looking at white's pawns on c3 and c4 they must have made 3 captures -
white's only catures.
And the last move could have been a pawn capture.
Apart from this I'm stuck, but I'm working on it!
<mik...@unsafe.nu> wrote in message news:3bdff709...@news.chello.se...
Glenn C. Rhoads
White's last move could NOT have been Qb6-a7 blocking the check
because then black would have no "unmove" available -- the white
king is in check and would still be so in every possible preceding
position.
First it helps to write down as many facts as you can deduce.
A) Black is missing 2 knights and one pawn.
B) White is missing 3 pawns.
Fact 1: White made three pawn captures to get his tripled c-pawns
and these are the ONLY captures white made (because of A)
C) Black made a pawn capture to get his doubled c-pawns.
D) This could not have been a capture of an unpromoted pawn
because white would need some additional pawn captures to get a
pawn in place (see Fact 1)
The above implies,
Fact 2: White promoted at least one pawn and it was captured
on either c6 or c5 by black's d-pawn (the a-pawn is at a7 and
the e-pawn is unmoved).
E) Black's missing pawn had to have been either an f or a g pawn.
F) for the missing pawn to have been captured, this pawn would
have to have made at least 3 captures to get to the c-file
where white made all his captures.
G) Black made at most three captures (from B) and two of these
were made to get the doubled c and e pawns leaving only one
more capture.
Thus,
Fact 3: Black promoted either his f or g pawn (from E,F,G) and
this was captured on either c3 or c4.
So what was the last move?
We've already ruled out Qb6-a7. Could it have been Nd5 takes
on c7? No because of Fact 1. Could it have been d or f takes
on e6? No again because of Fact 1. If white moved last it could
only have been either d2xc3 or d3xc4. In either case the captured
piece was either a knight or the promoted pawn.
I don't see how the previous moves are going to be unique.
If you undo any of the possible captures, then black's preceding
move could have been Qb6-b5. Then the previous white moves could
have been to shuffle the white rooks back and forth while the
resurrected black piece dances all over the place.
Also, there is nothing in the original position that suggests
white made the last move. I see three possible last moves for
black -- Rb6-a6, Qb6-b5, or Ke5-e6 (preceded by dxc4+).
Andrew Buchanan
the most notorious retro problems in existence, which would require extreme
tenacity to solve by hand. After the usual muttering about pawn captures,
the first phase of the solution involves shuffling the 7 units trapped in
the 4x2 rectangle, until a pawn move can happen. There are at least 102
single moves since that moment, so 50-move-rule issues apply. But only the
last 96 are unique.
You can find a solution at
http://www.softdecc.com/pdb/search.pdb?expression=PROBID='P0000228'. Gerd
Wilts who manages that database might also share with you if you email him
details of his computer analysis. Note that the move sequence is taken way
back further, to a point where the position is fully unlocked, not just the
first pawn move.
This amazing position was by Gyula Breyer in 1922, corrected by Gerd Wilts
in 1922.
Cheers,
Andrew Buchanan.
"Alan Trzuskoski" <trz...@rpi.edu> wrote in message
news:3BDEE546...@rpi.edu...
Andrew Andkjar
(http://www.softdecc.com/pdb/search.pdb?expression=PROBID='P0000228') you
posted, but there is still two problems I see with the suggested
solution....
1) The solution on the above web page seems to include only 86 moves, where
the problem stated is said to require 96 unambiguos moves.
2) The solution's notation is quite strange, and I cannot make sense of
it... (the only columns that seem to be mentioned are 'a' and 'b' , and
strange uppercase 'T', 'D', 'L','S'.... I know that either I am interpreting
the notation incorrectly, or something here is just not right.
"Andrew Buchanan" <ans...@mediaone.net> wrote in message
news:ef8F7.62066$xe.15...@typhoon.we.rr.com...
Sylvain Leroy
"Andrew Andkjar" <and...@hotmail.com> a écrit dans le message news:
WpJF7.15299$Ne.26...@typhoon.nyroc.rr.com...
> I checked out the link
> (http://www.softdecc.com/pdb/search.pdb?expression=PROBID='P0000228') you
> posted, but there is still two problems I see with the suggested
> solution....
>
> 1) The solution on the above web page seems to include only 86 moves,
where
> the problem stated is said to require 96 unambiguos moves.
>
> 2) The solution's notation is quite strange, and I cannot make sense of
> it... (the only columns that seem to be mentioned are 'a' and 'b' , and
> strange uppercase 'T', 'D', 'L','S'.... I know that either I am
interpreting
> the notation incorrectly, or something here is just not right.
Maybe german: TDLS = RQBN
John H. Guillory
<d.j.h...@durham.ac.uk> wrote:
>The queen didn't capture a piece on a7. I assume white is playing up the
>board (white king is on a8 not h1). Also you never said whose turn to move
>it is.
>
>Looking at white's pawns on c3 and c4 they must have made 3 captures -
>white's only catures.
>
>And the last move could have been a pawn capture.
The only way I see the last move could be (But then again, I'm not
that great of a chess player either) would be the Pawn at B7 moving
from C8 to capture a piece. Telling what piece that was before or any
move beyond that would a bit interesting, I'd imagine from the partial
view of the board, it'd been a pawn.... But if this is a computer
chess game (as the screen appears), the pawn could have very well made
it to the end of the board and been promoted to a queen, then returned
back to the white side in a feeble attempt to protect the King that
failed.