[Retros] Retro Corner News

[Henrik Juel]

>From: mri_two at t-online.de (Mario Richter)

>One thing that confuses me:

>On the one hand Ceriani's famous KxQ-LM-problem is listed as the

>record holder for categories A,B and C; on the other hand some

>Type B records have less pieces as the corresponding Type C problems

>(e.g. K-, Type B: 4 pieces, Type C: 5 pieces).

>So what makes a record a record?

>And one last question to the experts:

>The Type B-list doesn't contain an e.p.-capture.

>Is such a thing impossible or has it simply not yet been found?

>greetings

>  mario


The basic record list is the type A positions. In some cases adding 
conditions may allow a position with fewer men. Traditionally the 
conditions considered have been: Specifying which camp has the move 
(type B) or allowing one camp to check (implying that the other camp has 
the move) (type C). So from a record point of view, a type B position is 
of interest only if it has fewer men than the best type A one, and 
analogously for type C.

There are hundreds, if not thousands, of retros, where you can deduce 
that last move was an ep capture. I'm not familiar with the records, but 
for type B, the Amelung 1897 classic is a good guess:
Kf5 Th8 Pe5f6h5 - Kh6 Pg5h7 (5+3=8)
White to move. Last move?
(The original stipulation was #2).

Henrik
--
Henrik Juel. Informatics and Mathematical Modelling, Building 321,
Technical University of Denmark, DK-2800 Kongens Lyngby, Denmark.
http://www.imm.dtu.dk/~hj  Phone +45 4525 3391  Fax +45 4528 1344