[Retros] Task - double check mate
allanb at gofree.indigo.ie
Sun Oct 1 16:08:13 EDT 2006
----- Original Message -----
From: "Noam Elkies" <elkies at math.harvard.edu>
To: <Nicolas.Dupont at math.univ-lille1.fr>; <retros at janko.at>
Sent: Sunday, September 24, 2006 2:49 AM
Subject: Re: [Retros] Task
> > In the web-site "France-Echecs", I proposed the following problem :
> > Construct the shortest possible exact proof game ending by a check-mate
> > given by a piece returning at home.
> What's the link?
> > Here are our records, the 2 sections depending on whether or not
> > some piece is captured during the game :
> There could also be separate sections for discovered, direct
> and double check (though of course not all pieces can do
> all of them).
As suggested by Noam, here is a double check-mate (with captures) given by
a piece returning to its home square. Both checks are required.
1.Pa2-a4 Pe7-e5 2.Ra1-a3 Ke8-e7 3.Ra3-e3 Ke7-d6
4.Sb1-c3 Kd6-c5 5.Pb2-b4 Kc5xb4 6.Bc1-a3 Kb4xa3
7.Qd1-b1 Pb7-b5 8.Qb1xb5 Bf8-b4 9.Ke1-d1 Pd7-d6
10.Kd1-c1 Bc8-f5 11.Kc1-b1 Bf5xc2 12.Kb1-a1 Pf7-f5
But there must be a shorter setting.
> > Knight with capture : not least [...]
> What does "not least" mean?
> > Rook without capture : 9 moves (Itamar Faybish)
> This at least is not optimal:
> 1 f3 h5 2 Kf2 h4 3 Qe1 h3 4 Kg3 Nf6 5 Qf2 Rg8 6 Kh4 Se4 7 g3 f5 8 Qg2 Rh8#
> (C+ Popeye very quickly), and one can probably go below 8.0
> with some more effort.
> --Noam D. Elkies
> Retros mailing list
> Retros at janko.at
More information about the Retros