[Retros] New Task: PG m.0 #n

andrew buchanan andrew at anselan.com
Tue Oct 10 07:29:49 EDT 2006


Hi Joost,

That's a very enjoyable and unusual composition - thanks for posting it. The
reasoning is not too complicated - the longest chain I think being that
which shows that Black's last move was not Bd7-e8.

It's unlikely this could be tightened to become a proof game, but that isn't
to detract from it.

When I suggested that one should look for an SPG-# in which every unit was
essential for the direct mate, I did wonder about having retro elements in
the direct mate. In a way, this is a "cheating" way to achieve that goal.
The units acquire their essentiality through the retro aspects of the
composition, which then are relevant for the forward problem through e.p. or
castling.

Rather, I would be looking for the units to acquire their essentiality in
the usual way through the forward aspects of the composition. This may be
quite hard to engineer, particularly for interesting direct mate content.

Cheers,
Andrew.

-----Original Message-----
From: retros-bounces at janko.at [mailto:retros-bounces at janko.at] On Behalf Of
Joost de Heer
Sent: 07 October 2006 14:31
To: andrew at anselan.com; The Retrograde Analysis Mailing List
Subject: Re: [Retros] New Task: PG m.0 #n


> This suggested to me a different task. Create a PG in m.0 moves which is

> also a sound direct mate in n (>= 2) moves. This appeals to me because I

> like positions in which both retro and forward play must be considered.


Not an exact proofgame, but:

Caillaud, Loustau, Rotenberg
Phenix 10.1998
4b2r/p1P4p/1q2PB1k/r1Q1S1pP/1p6/pP1R4/P3BP2/1K
Position after black's 23rd move. #2v

1. Na3 c5 2. b3 Nc6 3. Bb2 d5 4. c4 Bd7 5. Qc2 b5 6. O-O-O Qb6 7. cxd5
Rd8 8. dxc6 f5 9. c7 Kf7 10. Kb1 Be8 11. Nf3 Rxd2 12. g4 Rxe2 13. Rd3
Nf6 14. Bxf6 Re5 15. h4 Kg6 16. h5+ Kh6 17. Rh4 b4 18. gxf5 bxa3 19. Rb4
cxb4 20. Be2 Ra5 21. Ne5 e6 22. fxe6 Bc5 23. Qxc5 g5

1. Rh3? (2. Sg4#) Bg6 2. hg6# (but 1. .. Qe6!)
1. hg6ep! (2. Rh3#) Bg6 2. Sg4
Djurasevic cycle

This is an extension to:

Caillaud, Loustau, Rotenberg
Special Prize
Phenix 10.1988 (version 12.1994)
4b3/7p/4S2k/1R2S1pP/8/3R4/2K5/3B
#2v

1. hg6ep? (2. Rh3) Bg6 2. Sg4# illegal
1. Rh3! (2. Sg4) Bg6 2. hg6#
Djurasevic cycle

Joost




More information about the Retros mailing list