1...Kc6-d5[+wPc6] 2.Kd1-c2 Kd5-e4[+wPd5] 3.Kc2-b3[+bPc2] Ke4-d3[+wPe4] {
} 4.Kb3-a4[+bPb3] Kd3-c4[+wPd3] 5.b3-b2 c6-c7 6.c2-c1=S Kc4-c3[+wPc4] {
} 7.Sc1-a2 + Kc3*b2[wKb2->e1][+wPc3] 8.Sa2*c3[bSc3->b8][+bPa2] c7*b8=R[wRb8->a1] {
} 9.Ka4-b3[+bPa4] {0-0-0
}
1...Kc6-b7[+wPc6] 2.Kd1-e2 Kb7-c8[+wPb7] 3.Ke2-f3[+bPe2] Kc8-d7 {
} 4.Kf3-g4[+bPf3] Kd7-e6[+wPd7] 5.Kg4-h5[+bPg4] Ke6-f5[+wPe6] 6.e2-e1=R e6-e7 {
} 7.Re1*e7[bRe7->h8] Kf5*g4[wKg4->e1][+wPf5] 8.Rh8-a8 b7*a8=R[wRa8->h1] + {
} 9.Kh5-g4[+bPh5] {0-0
(C+ by Popeye 4.75)}
This was inspired by a problem of Andreas Thoma, published in feenschach 205, I-II/2014, No. 10915:
wKg5, bKf2 (1+1)
h008, b) wKg5->g6
Anticirce, Circe, Sentinelles, Black Maximummer, White Maximummer (C+ Py 4.75)
a) 1.Ke1 [+bPf2] Kf6 [+wPg5] 2.Kd2 Kg7 [+wPf6] 3.Kc3 [+bPd2] Kh6 [+wPg7] 4.Kd4 [+bPc3] f7 5.Ke3 [+bPd4] f8R 6.d1Q Rxf2 [wRf2->a1; +bPf7] 7.Qh5+ Kxh5 [wKh5->e1; +bQd8; +wPh6] 8.Qa5 0-0-0,
b) 1.Kg3 [+bPf2] Kf7 [+wPg6] 2.Kf4 [+bPg3] Ke6 [+wPf7] 3.Kg5 [+bPf4] Kd5 [+wPe6] 4.Kh4 [+bPg5] Ke4 [+wPd5] 5.f1S Kf5 [+wPe4] 6.Se3+ Kxg5 [wKg5->e1; +bPg7; +wPf5] 7.Sxf5 [bSf5->g8; +wPf2; +bPe3] fxg8R [wRg8->h1, +bSg8]+ 8.Kg5 [+bPh4] 0-0.
When I saw this problem for the first time I thought that it is a wonderful idea, but after the second sight I wondered why the author used Circe. In my eyes Circe here is a triple weakness: 1. there is no need, because Anticirce is enough to get castling right for White, 2. it produces some more unnecessary Sentinelles-pawns and 3. the b)-solution works also without Circe, here Circe is just a cookstopper.
Therefore I tried to find a better presentation of this idea. To gain this I had to change everything: the position, the stipulation, the solution, and of course the fairy conditions, i.e. no Circe. I think my problem is original enough.
Although the solution is a halfmove longer, there are much less Sentinelles-Pawns in the final positions, the motivation for the black promotion is harmonious now, no Circe and I could even drop the twinning. I asked Andreas, whether he would prefer being co-author or only mentioned as "after Andreas Thoma". He meant that "after" would be enough. (Author)
Author’s comment is just added to the problem.