[Retros] series SPG

[Goran Wicklund]

Nicolas Dupont wrote:


> Dear retro friends;

> 

> In Richard's very interesting list of series mates, the first unknown

> feature is :

> 

> >Ng1: ?

> 

> I propose the following (one-sided) game :

> 

> 3) Ngb5 4) d4 5) Bg5 6) f4 9) Kh4 10) g3 11) Bg2 15) Re6 16) e3 17) Qf3 18)

> Qxb7

> 19) Qxc8 20) Nxc7 #

> 

> Of course, I'm unable to prove that 20 is  the shortest number to do that

> job. I'm even unable to be sure that this game is sound (one-way ticket).

> Indeed, "popeye" says nothing after 2 days of analysis, although it seems

> that everything  is OK :

> 

> The last position of the Ng1 (before mating) cannot be e6 because of the

> rook, cannot be d5 because the queen goes through that point, and cannot be

> reached to b5 via e2 because e3 comes too late. Moreover, b7 and Bc7 must

> have been captured by the queen, because there are only 20 moves.

> 

> So, what do you think about that ? I forgot something ?

> 


I am sure that this is correct and do not think that it can be done
in less than 20 moves. I was very close to find exactly the same 
game, and have tried many possible games for this task, but did not find 
any other dual free.

Here are three other open tasks with my proposed solutions:


Pc2: 4.b*c7 5.c*b8B 7.Bg5 8.f4 11.Kh4 12.g3 
     14.Bg4 15.Nh3 19.R*c8 23.c*d7#
     
Pd2: 1.h4 4.R*d7 6.Re4 9.d6 11.Q*b7 12.Q*b8
     13.Q*c8 14.d7#
     
Pc2->B: 4.c*b7 5.b*a8B 7.Bg6 8.e4 9.Qh5 10.f3 13.Kh4
        14.g3 16.Bf5 17.Nh3 21.R*c7 23.Re6 24.B*f7
        
Göran Wicklund