|
| Page: [Previous] [Next] 1 2 |
| (21) Posted by Hauke Reddmann [Wednesday, Aug 20, 2014 12:45] |
Aaaaand here is the addendum!
HR after Karan:
c) Nd6 etcetc., Black wins->White wins after *mirroring* (horizontally).
Nalimov-checked. Again unique (?!?).
Hauke
|
|
| (22) Posted by Arno Tungler [Thursday, Aug 21, 2014 08:44]; edited by Arno Tungler [14-08-21] |
I found one with 5 units. Hardly unique but I did not yet find a second...
Thus: On the empty board add minimum units [EDIT: NO pawns] to a position with White to move that is won. When flipped/horizontally mirrored/rotated by any/... it is lost.
Will give you some time to find. The first becomes co-author with Hauke and me...
PS: However, IF unique someone smarter should have already found it.
|
|
| (23) Posted by Geir Sune Tallaksen Østmoe [Thursday, Aug 21, 2014 11:06]; edited by Geir Sune Tallaksen Østmoe [14-08-21] |
 (= 3+2 )
I think there is a good chance this one is unique. All the similar positions I could think of were drawn without the castling.
|
|
| (24) Posted by Arno Tungler [Thursday, Aug 21, 2014 11:20] |
Wow, that was fast... But really unique??
|
|
| (25) Posted by Hauke Reddmann [Thursday, Aug 21, 2014 12:19] |
I think so. Ke1 Ra1 Ne3 - Kh1 Qe5 is drawn by Ra3 (too bad!)
It shouldn't be that difficult to check all positions - it can
only motivated by castling; since it's win/lose we can fix
White castling; with 5 pieces, castling can't mate immediately
with White having no more pieces, thus we can assume White has
K, R and a light piece and Black naturally a Q. Further the
castling must check and that on the first row (otherwise a
simple R check would suffice) - and the Q getting forked (or
maybe skewered) off immediately. Ke1 Ra1 Sf3 - Kh1 Qg2 doesn't
work either. Neither Ke1 Rh1 Sd3 - Kb1 Qa2.
You can check by hand now :-)
|
|
| (26) Posted by Geir Sune Tallaksen Østmoe [Thursday, Aug 21, 2014 12:36] |
Yes, that is very similar to my reasoning. With the limited possibilities we have, it shouldn't be too difficult to check all the possible candidate positions.
There is an alternative possibility, but I didn't find anything that was really close to working there: White could have a piece on c2 or g2, and it is crucial that by castling, White gives check AND covers the piece. Something like this:
 (= 3+2 )
But there are two problems with this position. 1.Rd2+ does not lose, and 1.0-0-0+ does not win. :)
|
|
No more posts |
Page: [Previous] [Next] 1 2
MatPlus.Net  Forum Promenade An almost joke question |
ISC 2024
