﻿﻿ MatPlus.Net

Website founded by
Milan Velimirović
in 2006

22:32 UTC
 ISC 2024

Remember me

 CHESS SOLVINGTournamentsRating lists1-Jul-2024
 B P C F

MatPlus.Net Forum General Construct the longest ser-x with a Black homebase

Page: [Previous] [Next] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

### Construct the longest ser-x with a Black homebase

Minor Edit 4227/2022: I have slightly altered the title to reflect the nature in which this thread has gone.

Tis' was a session messing around with the pieces, when an idea popped up. Create a legal seriesmover selfmate as long as possible, but all Black pieces have to be at homebase. This heavily restricts the possibilities for both sides. White pawns cannot be doubled, and too many can allow for cooking promotions. The most viable Black mates are Qxc7, forced by a knight; Bxg7, also forced by a knight; or c(e)xd6, again forced by a knight.

One must use extensively the geometry of the board to create a flawless plan of action. Multiple possible mating positions exist that can ruin a problem. Even c1 is a possibility in some circumstances.

For those who wish to, I challenge you to construct the longest ser-s#n with a Black homebase.

I start the auction with a bid of ser-s#9 (not C+, but simple logic).
(= 4+16 )

1. Nxg8 2. h6 3. hxg7 4. Nh6 5. g8=B 6. Bxh7 7. Bb1 8. Nf5 9. Ng7+ Bxg7#

EDIT: Here is a schema for a ser-s#16 for further inspiration, as it is no doubt busted.
(= 7+16 )

1. Rd4 2. Kd2 3. Ke3 4. Kf4 5. Ke5 6. Rf4 7. g4 8. g5 9. g6 10. gxh7 11. hxg8=N 12. Nh6 13. Ng4 14. Ne3 15. Nc4 16. Nd6+ c(e)xd6#

Yes, it's busted as the white king can go to e5 through d4 as well, so you need to drop the first move. Furthermore, what works on the west works on the east: hxg8=S Sf6 e(g)xf6#

With Kh2 and Rd2-d4, at least some moves can be saved?

Possibly then also Rd4=Pd4

Some possible schemes:
(= 5+16 )

ser-s#12 (C?)

And your last scheme can perhaps be modified:
(= 7+16 )

ser-s#15 (C?)
EDIT: No, cooked with Sxc7 Qxc7 with pieces blocking the king on d5, d4, e4 and f5.

@Siegfried: Pd4 doesn't work (illegal double pawn)

[EDIT Previous diagram cooked, new attempt]
(= 4+16 )

ser-s#14 (C+)
1. a4 2. a5 3. a6 4. ab7 5. ba8=S 6. Sb6 7. Sd7 8. Se5 9. Sg4 10. Ke5 11. Bf5 12. Sh6 13. Sg8 14. Sf6 e/gxf6#
Fully tested by Olaf Jenkner.

This doesn't work:
(= 4+16 )

ser-s#15
Cook: 5. ba8=S 7. Sd7 8. Kf5 9. Rh4 10. Bf4 11. Kg5 12. Rh7 13. Rg7 14. Rg8 15. Sf6 ef6#

3 solutions in 13: 1.Bf4, 1.Kf4, 1.Kh4, all ending with 13.Sd7-f6+ e,g:f6#

@Frank already found those cooks, changed the diagram.

I see ... nice idea and position.

First diagramm in post (5) is C+ .
Interesting cook: 1.Rf1 2.a4 3.a5 4.a6 5.axb7 6.bxc8=Q 7.Dxb8 8.Dxc7 9.Rd1 10.Txd7 11.Kd5 12.Kc6 13.Rd5 14.Rc5 15.Qd7 Qxd7#

Well done, Joost! The fewer the units, the greater the likelihood of soundness. If only wK+3 units can produce 14 moves already, perhaps we could also hunt by piece count. wK+1 unit is no doubt dualed, if not outright impossible. For wK+2 units, I present ser-s#7 to start--I couldn't manage an excelsior for more moves.

(= 3+16 )

Olaf, if you please, is my ser-s#9 C+?

QUOTE
Olaf, if you please, is my ser-s#9 C+?

C+ with Popeye in 12 seconds on my old laptop.

Thanks to Olaf's cook:
(= 3+16 )

ser-s#11 (C+)
1. Rc7 2. Rc8 3. Rb8 4. Rb7 5. Rd7 6. Kc6 7. Rd5 8. Rc5 9. d5 10. d6 11. d7 Qd7#

And two fairy variations:
(= 3+16 )

ser-s#14 (C+, Berolina pawn on c2)

(= 3+16 )

ser-s#15 (C+, swapped Ra8 and Bc8)

Great stuff, James & Joost.

For Joost's latest orthodox version, what happens if one retracts one move further? E.g. 0. Kc4-d5 or 0. Rh1-c1?

Rh1-c1: dual 1. Rb1 2. Rb7 3. Rb8 4. Rc8 5. Rc7 6. Rd7
Kc4-d5: dual 1. Kb5 etc, or 1. d5 2. d6 3. Kd5 and the rook can go to c5 through c7.

(= 3+16 )

ser-s#12 (C+)

(= 3+16 )

ser-s#17 (C?)

The intention is awesome: 4. hg7 5. gh8=R 6. Rh7 7. Rf7 10. Rd7 11. Ke6 12. Qb7 13. Qb8 14. Qc8 16. Re5 17. Qd7 Qd7#

But alas cooked:
5. gxh8=S 6.Qh2 7. Qc7 8. Qb7 9. Sg6 10. Se7 11. Sc6 12. Sb8 13. Sd7 14. Kc6 15. Sb6 16. Sc8 17. Qd7 Qxd7#

You can use Jacobi to search for schemes, e.g. I found this cook with the following Jacobi input while trying to see if a knight promotion would work:
fors rsbqkbsS/pppppp1p/8/3K4/8/8/8/7Q
stip ser-*a=>b11 pie whi kc6 addpieces
stip s#1

Joost, thanks for checking this. Tough cook. Not sure if I'll manage to fix this problem but I'll try it.

Does this work?
(= 3+16 )

ser-s#15 (H+ with the assumption that the king is mated on c6, no solution with the white king on e6 in the final position)
Unfortunately you can't add 1. Qg3-d6 as first move, because that's definitely cooked.
1. g4 2. g5 3. g6 4. gh7 5. hg8=S 6. Se7 7. Sc8 8. Sb6 9. Sd7 10. Sb8 11. Sc6 12. Sa5 13. Sb7 14. Kc6 15. Qd7 Qd7#

Congratulations, C+