Die Schwalbe

6 problem(s) found in 7575 milliseconds (displaying 6 problem(s)). [COMMENTDATE>=20200919 AND NOT K='Hilfsrückzüger' AND NOT S='Die Welt' AND A='Rol, Guus'] [download as LaTeX]

1 - P0008413
Guus Rol
Harry Goldsteen

Probleemblad 1 01-02/1989
P0008413
(8+12)
#2 durch Schwarz
1. Kxc2 Sb3 2. ... Tc1#

Bsp-Auflösung Zvi Mendlowitz(PDB 2022-09-19)
R: 1. ... Ta2-a1 2. a3xLb4 Lf8-b4 3. Lb4-c3 Kb3-a4 4. Le7-b4 a7xSb6 5. Sa4-b6 Dd5-d4 6. Sc3-a4 Dd4-d5 7. Sb1-c3 Ta1-a2 8. Lf6-e7 Kb4-b3 9. a2-a3 Kc5-b4 10. Lg5-f6 Kd6-c5 11. Lf6-g5 Kd7-d6 12. Le7-f6 Kc8-d7 13. Lf6-e7 Kb8-c8 14. Ld8-f6 e7-e6 15. d7-d8=L Kc8-b8 16. e6xTd7 Td8-d7 17. f5xLe6 0-0-0 18. f4-f5 Lc8-e6 19. f3-f4 Dd8-d4 20. f2-f3 d7xLc6 21. Lf3-c6 Sc6-a5 22. Le2-f3 Sb8-c6 23. Lf1-e2 Sc6-b8 24. e2xSd3 Sb4-d3 25. Kd1-c1 d3xDc2 26. Ke1-d1 e4xTd3 27. Dd1-c2 f5xDe4 28. Sa3-b1 Tc1-a1 29. Dh4-e4 Tc3-c1 30. Dh8-h4 Tb3-c3 31. Tc3-d3 g6xTf5 32. Tc1-c3 Tc3-b3 33. Ta1-c1 Tc1-c3 34. h7-h8=D Tc3xLc1 35. h6-h7 Tg3-c3 36. Th5-f5 Sb8-c6 37. Th1-h5 Sd5-b4 38. h5-h6 Sf6-d5 39. h4-h5 Sg8-f6 40. h2-h4 Th3-g3 41. c2-c4 Th8-h3 42. Sb1-a3 h7xSg6 43. Sh4-g6 Sc6-b8 44. Sf3-h4 Sb8-c6 45. Sg1-f3
play all play one stop play next play all
"O, W, thuis..."
Henrik Juel: 1.Kxc2 Sb3 2.. Rc1#. -1.. Ra2 -2.a3:B Bf8 -3.Bb4 Kb3 -4.Be7 a7:S -5.Sa4 Qd5
-6.Sc3 Qd4 -7.Sb1 Ra1 -8.Bf6 Kb4 -9.a2 Kc5 -10.Bg5 Kd6 -11.Bf6 Kd7 -12.Be7 Kc8 -13.Bf6 Kb8 -14.Bd8 e7 -15.B=d7 Kc8 -16.e6:R Rd8 -17.f5:B 0-0-0 -18.f4 Bc8 -19.f3 Qd8 -20.f2 d7:B -21.Bf3 Sc6 etc. (2003-08-26)
Yoav Ben-Zvi: An attempt at a coherent argument that reveals the intricate sequence of events in the resolution while refuting substantial alternatives:
The promotion of wBc3, on d8 or f8, requires 2 captures by the promoting wP. These, together with the captures by wPd3 and wPb4, account for all 4 missing Black pieces. Black captured [wBc1] at home. The other 7 missing White pieces were captured by bPs. [wBf1] cannot be released (by wPe2xd3) until [bPh7] plays bPd3xc2 completing the capture of 5 White pieces. Therefore [wBf1] was captured by bPd7xBc6 (the remaining capture by Black on a light colored square), releasing [bBc8]. This means that [wBf1] is released (by wPe2xd3) before [bBc8] is released (by capturing the previously released wB on c6) so the bB could not have been the piece captured on d3. It also could not be captured on b4 (dark colored square) so it was captured by White's promoting pawn. This means the promotion could not have been played by capture from e7 as that would imply that the 2 captures by the promoting pawn were both on dark colored squares in contradiction with the requirement to capture [bBc8]. Therefore the promotion move was wPd7-d8=B with bK standing outside the squares c8,d7,d8,e8.
Looking backward: To avoid White retrostalemate the first 2 retractions must be -1...bRa2-a1+ -2.wPa3xb4.
After this, unlocking the South-West cage and extracting bRa2 requires retraction of wPe2xd3 which requires previous uncapture of wB by bPd7xBc6, so it can be retracted home to f1. bP standing on d7 implies that a bB must previously be uncaptured and retracted home to c8, preceded by retracting BPe7-e6 to clear the path of uncaptured bB to c8. Retreat of bP to e7 is preceded by uncapture and return home of [bBf8]. With both bBs locked at home and bP on e7 the Black Central cage is locked by retracting bPd7xBc6 so bK and bQ must also be retracted home first. It follows that bKa4 needs to exit the West cage before it has been unlocked (to get back home to e8 before the Black Central cage is locked). The bK cannot exit via b5 as this implies retraction of wPc3-c4+ but, with c3 occuppied by wP, bRa2 does not have a valid exit path (not via c3 as it is obstructed by wP or via f1 as it will be occupied by wB). wP standing on c4 prevents retracting bN away from a5 so bK cannot exit via a5. The only valid exit of the bK is via b4 implying retracton of wPa2-a3 which requires bRa2 to retract to a1 (vacating a2 for the wP) which requires provision of a shield, standing on b1, to protect wKc1 from bRa1.
After the unpromotion of wBc3, White has only pawn retractions left to make until a White officer is uncaptured. This is not enough to allow Black to complete the retro-maneuvers required in preparation for retracting bPd7xBc6 (getting bBs,bK,bQ home). It follows that, at some point in the middle of its preparation maneuver, Black will need to uncapture a White officer which provides the tempo retractions needed to complete the maneuver. This is only possible by bPa7xb6. With bP standing on a7, [bRa8] cannot be retracted home via the a file so it must retract to its original corner via row 8 before the bB is locked at c8. This implies that [bRa8] is uncaptured by a wP prior to unlocking of the cage.
Uncaptures by White prior to the unlocking of the South-West cage (one available on b4 and two by pawn unpromoted from wBc3) include uncaptures of both bBs and a bR since, as noted in the above analysis, all 3 must be uncaptured and retracted home before retracting bPd7xBc6 (all of which precedes the unlocking of the South-East cage). This means that the bN is uncaptured later (on d3) so it cannot provide the shield on b1. The shield also could not be provided by [bBc8] as this bishop must be returned home when it is needed to provide the shield. The only remaining possibility is that the shield on b1 is provided by a wN that is uncaptured by bPa7xNb6. The wN on b1 is immobile until the wK can be retracted away from c1 so here too, once the unpromotion occurs Black must perform the preparations for retraction of bPd7xBc6 (which now includes return of the uncaptured bR to a8 or b8) under time pressure, completing the required maneuvers before White runs out of tempo retractions. Having determined the structural elements it is not too difficult to construct a resolution that succeeds in preparing for retraction of bPd7xBc6 before White runs out of tempo retractions. The resolution maintains the following: delay the unpromotion (known from analysis above to be wPd7-d8=B) while retracting bK to b8; bQ to d4 and bB (uncaptured on b4) to f8 and continuing by retracting -1.wBd8-? Pe7-e6 -2.Pd7-d8=B Kc8-b8 -3.Pe6xRd7+ Rd8-d7 -4.Pf5xBe6 0-0-0 -5.Pf4-f5 Bc8-e6 -6.Pf3-f4 Qd8-d4 -7.Pf2-f3 Pd7xBf6 -8.Bc6-? Nc6-a5. Following this the wB is retracted to f1 allowing the unlocking of the cage by wPe2xNd3 and bRa1 is retracted back home to h8 via c1,c3. One of the White pieces uncaptured by bPc2 unpromotes on h8. The alternative of uncapturing the bR on b4 (instead of bB) fails to complete the preparations on time.
To draw attention to the rich, unconditional, content of the resolution the stipulation could be "Squares that must have been occupied by bRa1". This problem deserves a full SPG. The best I can manage is a game of 53.0 moves. (2018-12-23)
Zvi Mendlowitz: Proof game in 44.0 moves:
1. Nf3 Nc6 2. Nh4 Nb8 3. Ng6 hxg6 4. Na3 Rh3 5. c4 Rg3 6. h4 Nf6 7. h5 Nd5 8. h6 Nb4 9. Rh5 N8c6 10. Rf5 Rc3 11. h7 Rxc1 12. h8=Q Rc3 13. Rc1 Rb3 14. Rc3 gxf5 15. Rd3 Rc3 16. Qh4 Rc1 17. Qe4 Ra1 18. Nb1 fxe4 19. Qc2 exd3 20. Kd1 dxc2+ 21. Kc1 Nd3+ 22. exd3 Nb8 23. Be2 Nc6 24. Bf3 Na5 25. Bc6 dxc6 26. f3 Qd4 27. f4 Be6 28. f5 O-O-O 29. fxe6 Rd7 30. exd7+ Kb8 31. d8=B e6 32. Bf6 Kc8 33. Be7 Kd7 34. Bf6 Kd6 35. Bg5 Kc5 36. Bf6 Kb4 37. a3+ Kb3 38. Be7 Ra2 39. Nc3 Qd5 40. Na4 Qd4 41. Nb6 axb6 42. Bb4 Ka4 43. Bc3 Bb4 44. axb4 Ra1+ (2022-09-19)
comment

Genre: Retro
FEN: 8/1pp2pp1/1pp1p3/n7/kPPq4/2BP4/1PpP2P1/r1K5
Input: Gerd Wilts, 1996-09-17
Last update: Mario Richter, 2022-09-19 more...
2 - P1096298
Guus Rol
R362 Probleemblad 10-12/2009
P1096298
(6+3) C+
h#2.5
2 Lösungen
Circe (AP)
1. ... dxc6ep[+sBc7] 2. 0-0-0 c5 3. Tg8 Dxg8[+sTa8]#
1. ... Kc6 2. Txa4 Dxa4[+sTa8] 3. 0-0-0 Da8#
play all play one stop play next play all
1. ... dxc6ep[+sBc7] 2. Txa4? Dxa4[+sTa8] 3. 0-0-0 Da8# doesn't work because we don't know whether Ta8 in the diagram position had already moved.
There are another 11 retro tries where Black does not castle at all, so can be eliminated in the usual way.
Henrik Juel: 1...d5*c6ep[+bPc7] 2.Ra8*a4 Qa2*a4[+bRa8] 3.0-0-0 Qa4-a8#
1...Kb5-c6 2.Ra8*a4 Qa2*a4[+bRa8] 3.0-0-0 Qa4-a8# (2022-01-05)
A.Buchanan: I think that the solution with e.p. is 1. ... dxc6ep[+sBc7] 2. 0-0-0 c5 3. Tg8 Dxg8[+bTa8]# The one that you give Henrik doesn't work in Circe, because we don't know whether sTa8 had moved there prior to the diagram position (2022-01-07)
more ...
comment
Keywords: a posteriori (AP) (Type Petrovic), Circe, Castling, En passant as key
Genre: Retro, Fairies, h#
Computer test: HC+ Popeye 4.61 + thinking about AP
FEN: r3k3/8/8/1KpP4/P1P5/8/QP6/8
Input: Gerd Wilts, 2009-12-19
Last update: A.Buchanan, 2022-03-21 more...
3 - P1380353
James Malcom
Guus Rol

www.chess.com 15/7/2020
P1380353
(11+11) C+
#4
1. d8=L!
1. ... Tb8 2. cxb8=L! d4 3. Le5 Lxf4+ 4. Lxf4#
1. ... Txd8 2. cxd8=S! d4 3. Se6 Lxf4+ 4. Sxf4#
1. ... Ta8 2. h8=D! Txd8 3. Txe3+ Kxe3 4. De5#
2. ... d4 3. Dxd4 ... 4. Dxe3#,Txe3#
1. ... Txc7 2. Lxc7 d4 3. Le5 Lxf4+ 4. Lxf4#
1. ... d4 2. Sc4 ... 3. Txd2+ Ke1 4. Lxf2#
play all play one stop play next play all
https://www.chess.com/forum/view/more-puzzles/a-challenging-mate-in-5-can-you-find-the-first-move?page=3
Henrik Juel: C+ Popeye 4.61 (except for a mate dual) (2020-09-27)
more ...
comment
Keywords: under-promotion key (L), Promotion (D), under-promotion (S)
Genre: n#
Computer test: Popeye 4.61
FEN: 2r1R3/2PP3P/1N5K/1P1p4/1p3P2/1B1pbp2/3pkrp1/3R1bB1
Input: James Malcom, 2020-09-27
Last update: A.Buchanan, 2020-10-13 more...
4 - P1381928
Guus Rol
www.chess.com 14/11/2020
P1381928
(4+5) C+
#1
a) 1. Df5#,De6#,Df7#? aber Schwarz ist am Zug!
a) 1. ... Kxg6 2. Df5#
1. ... Lxg4 2. Df7#
1. ... Lxg6 2. De6#
play all play one stop play next play all
Henrik Juel: C+ Popeye 4.61
missing variation: 1. ... Kxg6 2. Df5#
If White had the move, 1.Df7,De6,Df5# would mate immediately
But Black has no last move, so he has the move, and his three possibilities are met with just these mates (2020-11-15)
comment
Keywords: No legal last move for Black, Fleck Theme (3)
Genre: Retro
Computer test: C+ Popeye 4.61
FEN: 8/6p1/3K1kPb/3Q2pb/6P1/8/8/8
Input: A.Buchanan, 2020-11-15
Last update: Alfred Pfeiffer, 2020-11-17 more...
5 - P1382697
Guus Rol
MatPlus.net Forum 29/11/2020
P1382697
(4+7)
What's the best White can do?
b) sTg1
a) No permitted moves
b) 1. fxg6ep+ Kxg6 2. Dg5#
play all play one stop play next play all
The end position is dead if Black's last move was R: 1. ... g6-g5+ (stalemate is inevitable) but not if it was R: 1. ... g7-g5+ which allows for e.p. Looking at the position prior to the last move:
a) the position with sB on g6 is not dead regardless since the sD can intervene instead of the bP move. This makes the e.p. invalid which would force 1. Dxg5+ Txg5=
b) the position with sB on g6 is already dead so the sB must have come from g7 allowing white to win by 1. fxg6ep+
AB: So what happens in a)? We can't know from the position, even with DP. If White can ep, he wins; if he can't the position is newly dead but still legal (as the alternative last move Qe3 would have kept the game alive). The ep convention specifically resolves the permissibility of the ep itself, but it's too narrow to claim (as at one point I did) that it fails to resolve the permissibility of the contingent Qg5+. So there are no permitted moves.

Suggested position evaluation in orthodox problem chess (OPC) (sketch):
1) What legal moves are there? ("legal" as per FIDE)
2) If 0 legal moves, then is there check? Yes/No => Mate/Pat.
Otherwise there is at least 1 legal move.
If an ep is legal set flag
3) Has this position occurred before? Yes/No => 3Rep/continue.
Is this RETRO? Yes/No => continue/play any legal move.
4) 50.0 moves since last capture or pawn move? Yes/No => 50M/continue
5) Check the child nodes recursively until find which mate. Success/Fail? => play any legal move)/Dead
Henrik Juel: 1.Dxg5+ Txg5= seems to secure half a point in both twins
So there is probably some DP trickery, which Andrew will explain soon... (2020-12-03)
Yoav Ben-Zvi: The end position is dead if Black's last move was -1...bPg6-g5 (Stalemate is inevitable) but not if it was -1...bPg7-g5 which allows for e.p. Looking at the position prior to the last move: in a) it is not dead regardless since the bQ can intervene instead of the bP move (this makes the e.p. invalid forcing 1.wQxg5) but in b) the position with bP on g6 is already dead so the bP must have come from g7 allowing white to win by 1.wPf5xg6 e.p. (2020-12-04)
A.Buchanan: Suggested principle for orthodox problem chess (OPC): playing forwards chess should always be on the basis (at least conceptually) of knowing the full gamestate (castling, e.p.). One can then run the forwards chess for each combination. Call this a "scenario" Some combinations can be rejected as illegal, and we can then hand over to the castling/ep conventions to clear things up.
In a), this results in game over. This makes sense, because under DP the drawing 1.Qg5+ is only playable if the living 1.fxg6ep+ is also playable. So the ep convention nixes both.
And how does the game end exactly? The same thing can happen in a simpler position: 8/k7/2p5/KpP5/1P6/8/8/8 WTM. There are two retro-possibilities, one is pat and the other keeps on going. It's nothing to do with DP rule.
Under RS: The ep convention ends the game, but we can't assume that the history of the game was not the double pawn move! We are just not *permitted* to make this move. We have "draw by ep convention".
Under PRA: our solution comprises one twin, and in that twin, we can assume that ep is actually illegal (or it should be if the conventions were worded properly!) and the position is pat.
The goal here is to codify OPC properly so that it can serve as foundation for retro fairies.
OPC allows for the variations in the castling/ep conventions & protocols. It merely hands off the results of each gamestate scenario. (2022-01-19)
more ...
comment
Keywords: En passant, Forsberg Twins, Dead Position
Genre: Retro
FEN: 6r1/5p1p/5P1k/5Pp1/6pK/8/3Q4/6q1
Input: A.Buchanan, 2020-12-03
Last update: A.Buchanan, 2022-01-19 more...
6 - P1390561
Guus Rol
Andrew Buchanan

Discord Chess Problems & Studies Server 02/05/2021
5th Place
Discord Composition Tourney 1
P1390561
(15+9)
#1
1. Dd2!#
1. 0-0-0?#
R: 1. Kc3-d3 De4-b4+ ( 2. Kd3-c3 Db4-e4 retro loop)
2. Kb2-c3 Lb4-a3+ ( 3. Kc3-b2 La3-b4+ retro loop)
3. Kb1-b2 Ta2-a1,Ta3-a1+
play all play one stop play next play all
Authors' solution: hey wait: so is there a dual? Well no. What was Black's last move? Let's look at all the captures & promotions first:
White: captured P promoted RRRRRBS
Black: captured QRRBBSS promoted -

bPs all on original columns: no black cross-captures. So 7 wPs captured once and promoted. The exception must have been poor wPh, because it can't get past bPh anyway, and was waylaid on its starting file by an officer. Other wPs captured as axb, bxc, cxd, dxc/e, exd, fxe, gxh before promoting.

The sequence of bP moves must have been one of:
(i) bPc6 (allows wPd to pass) then bPd5 (allows promoted rooks to escape) then bPe6 (when all the rooks are out, letting wPf to pass)
(ii) bPe6 then bPd5 then bPc6.

So which? In (i) the units promoting late on e8 & h8 must have been S & B, as the door was closed on escaping rooks. But e8 is a light square, and h8 is blocked by bPg7 which never moved. So neither could have promoted to B. Therefore it was option (ii), pawn move sequence roughly:
bPe6 & bPh6
gxh=R, fxe=R, dxe=R,
bPd5
exd=R, cxd=R
5 Rs all exit
bPc6
bxc=S
bPb6
axb=B

Bottom line: all the bP moves and captures were historical, and so it must be bK which just moved. We can now use Retractor 2 program from Theodore Hwa at http://xenon.stanford.edu/~hwatheod/retractor2/retractor.html to validate our manual analysis of the retro play, ignoring any suggestions which involve capture. So wRa had already moved: no castling now.
Henrik Juel: Black captured [Ph2] with an officer
The other white pawns captured all missing black men to promote seven times
Last move was not made by a black pawn as this would ruin the promotions, so it was Kc3-d3
Retroplay: 1... Kc3-d3 2.De4-b4+ Kb2-c3! 3.Lb4-a3+ Kb1-b2! 4.Ta3-a1+ or Ta2-a1+ (other K retractions just postpone the inevitable), so 1.0-0-0 is illegal
1.Dd2#
Good problem, maybe the best in the tourney (2021-06-08)
A.Buchanan: Thanks Henrik. Can you figure out which pawns promoted to which kind of officers on which squares? (2021-06-09)
Henrik Juel: No, I would expect there were several feasible promotion patterns
But the need to move [Pb7c7d7e7] very carefully can easily make your suggestion correct
Offhand I do not see the right sequence, so you are welcome to enlighten me (2021-06-10)
Henrik Juel: I think I got it
(promote [Pg2] to rook on h8 at some early time)
first e7-e6, promote [Pd2f2] to rooks on e8
then d7-d5, promote [Pc2e2] to rooks on d8
get all promoted rooks out
finally [Pb2] promotes to knight on c8 and [Pa2] to bishop on b8 (2021-06-10)
A.Buchanan: Exactly! If you try the other way round, there’s nowhere to promote B (2021-06-10)
comment
Keywords: Non-standard material (TTTTTLS), Cant Castler
Genre: Retro
Computer test: HC+ Popeye v4.85 Retractor v2.0 + thinking
FEN: 8/p4pp1/1pp1p2p/1RBpR3/RQ3RB1/BN1k4/2N3NR/R3K2R
Reprints: Facebook 28/05/2021
Input: A.Buchanan, 2021-06-08
Last update: A.Buchanan, 2021-06-11 more...
Show statistic for complete result. Show search result faster by using ids.

https://pdb.dieschwalbe.de/search.jsp?expression=COMMENTDATE%3E%3D20200919+AND+NOT+K%3D%27Hilfsr%C3%BCckz%C3%BCger%27+AND+NOT+S%3D%27Die+Welt%27+AND+A%3D%27Rol%2C+Guus%27

The problems of this query have been registered by the following contributors:

Gerd Wilts (2)
James Malcom (1)
A.Buchanan (3)