Website founded by
Milan Velimirović
in 2006

11:40 CET



Remember me

Forgot your
Click here!
to create your account if you don't already have one.

Rating lists


MatPlus.Net Forum Misc Construction Challenge
You can only view this page!
Page: [Previous] [Next] 1 2
(1) Posted by Bob Baker [Sunday, Apr 7, 2019 12:32]

Construction Challenge

The drawn position below is one I made in which the first 18 half-moves are "only" moves for both sides. I have been wondering if it is a record length for moves of this kind. As a way of finding out, I would like to issue the following challenge to the problem experts at MatPlus.Net:

Construct a legal drawn position with no promoted pieces in which the first 19 or more half-moves are the only moves each side could make to hold a draw.

If you accomplish this task, I can't guarantee you will become famous. But you will at least win my heartfelt admiration.

(= 6+6 )

(Read Only)pid=17158
(2) Posted by Steffen Slumstrup Nielsen [Sunday, Apr 7, 2019 19:35]

I just took a quick look at my own somewhat 50 draw studies. A couple of them begin with 10 only moves, but that is it. I guess much longer sequences with only moves do not happen by chance in a study. I will try to construct something :-)
(Read Only)pid=17159
(3) Posted by Hauke Reddmann [Sunday, Apr 7, 2019 22:38]

Well, I'm astonished that so many only moves occur in
a relatively "natural" position. (As OTB player I can
judge that.) Bob, just to be safe, have you run the
position through a computer?
(Read Only)pid=17160
(4) Posted by Geir Sune Tallaksen Østmoe [Sunday, Apr 7, 2019 23:17]

I agree, Hauke, that is very surprising.

My position looks much less natural, but I think it works. 25 half-moves.

(= 10+11 )

(Read Only)pid=17161
(5) Posted by Bob Baker [Monday, Apr 8, 2019 01:33]

Hauke, I had verified my position, and now I have also verified Geir's position. Geir is now the current record holder with a position whose starting "only" move count is 25 half-moves. I hope his admirable achievement will not discourage others from trying to increase the move count.

I suppose I could claim to be the record holder still for the top "natural looking" position, but that is really too subjective to be a meaningful task.

Do you know of anything ever having been published on this construction task?
(Read Only)pid=17162
(6) Posted by Steffen Slumstrup Nielsen [Monday, Apr 8, 2019 07:15]

I just realized that there are some known knight chases that must have the same characteristics.

For instance this by Popov. The moves are "only moves" beginning from 2. Be5+. I am sure there are longer ones?
(Read Only)pid=17163
(7) Posted by Bob Baker [Monday, Apr 8, 2019 08:03]; edited by Bob Baker [19-04-08]

Thanks, Steffen. This study made me realize I didn't represent my actual intent with the wording of my challenge. Here is my new challenge.

Construct a legal drawn position with no promoted pieces in which the first 19 or more half-moves are ! moves for both sides under the Nunn Convention.

The difference is that, under the Nunn Convention, a move doesn't get a ! if it is the only legal move. I hope this change doesn't make the task too artificial to be interesting. Geir's position doesn't survive the new wording, but I suspect he can quickly correct that if so inclined.
(Read Only)pid=17164
(8) Posted by Steffen Slumstrup Nielsen [Monday, Apr 8, 2019 09:10]

I like your clarification. That makes the task much less automatic, ruling out forcing checking sequences.

I agree that especially the later half (rook vs. pawns) of your own position is remarkable.
(Read Only)pid=17165
(9) Posted by Hauke Reddmann [Monday, Apr 8, 2019 09:33]; edited by Hauke Reddmann [19-04-08]

You beat me to it :-) I also suddenly remembered the
knight wheel and agree the record should only count under
Nunn convention.
I suggest yet another amendment...

(= 3+3 )

...or you might have to beat "infinity" or at least
a triple bonus :-)
(I.e., count will stop at first forced repetition of position.
In case Black can try yet another alternative instead of
immediate repetition, he may and it shall count.)

EDIT: Replaced invalid example.
(Read Only)pid=17166
(10) Posted by Geir Sune Tallaksen Østmoe [Monday, Apr 8, 2019 10:20]

I agree, ruling out forced moves is a good clarification, and it makes the task significantly more challenging. I will see what I can do with the adjusted task. I will probably have to rebuild the position, since all the first ten half-moves are based on the forced ...Kxc8.

I also assumed that perpetual checks don’t count as infinity, but it’s a relevant question whether you count all moves up to a threefold repetition, or whether you stop at the first repetition.

Another question is whether promotion duals are allowed.
(Read Only)pid=17167
(11) Posted by Bob Baker [Monday, Apr 8, 2019 10:21]

Thanks Hauke. My understanding of the Nunn Convention comes from the 2nd edition of Muller and Lamprecht's "Secrets of Pawn Endings. (It's one of my favorite books, and not just because they put my name in the preface for having sent them a correction.) They say that a move which just produces a repetition of the position does not get the exclamation mark, so an amendment may not be needed.

I appreciate the responses I've gotten. The advice I got at to bring my question to has turned out to be on target.
(Read Only)pid=17168
(12) Posted by Geir Sune Tallaksen Østmoe [Monday, Apr 8, 2019 10:25]

Oh, and another challenging task: Longest sequence without captures. The endgame with rook vs two pawns in the first post is an impressive start for that task.
(Read Only)pid=17169
(13) Posted by ichai [Monday, Apr 8, 2019 10:45]

Could someone give the solutions ?
(Read Only)pid=17170
(14) Posted by Bob Baker [Monday, Apr 8, 2019 11:13]; edited by Bob Baker [19-04-08]

Geir, there are cases when White is winning and can make a winning move that has to be immediately retracted to keep winning, and has merely added two moves to the winning line. I think that move is what Muller and Lamprecht were saying does not stop the real winning move from being a ! move. It leaves open the question of which are the ! moves in a perpetual check situation, a question that has to be answered if my challenge is not to be ambiguous. My inclination is to count all the moves up to the triple occurrence.

In the case of a promotion dual, I don't think the promoting move could be a ! move.
(Read Only)pid=17171
(15) Posted by Hauke Reddmann [Monday, Apr 8, 2019 12:41]

My gut feeling is we should *not* apply study conventions
(i.e. allow loss-of-time duals), and neither should Black
be able to try something else, since this is more of an
construction task.
BTW, I asked on Chess Stack Overflow whether someone could
come up with database examples (record or not) or lines
from the Nunn books.
(Read Only)pid=17172
(16) Posted by Bob Baker [Monday, Apr 8, 2019 12:52]

ichai, these are the moves for my position: (Nunn Convention) 1. Qxh3! Nxd3+! 2. Qxd3! Rxf5+! 3. Qxf5! Bxf5! 4. Kxf5! Rg8! 5. e6! Kb5! 6. e7! Kc6! 7. Ke6! Rh8! 8. d7! Rh6+! 9. Kf7! Rh7+! 10. Ke6= (10. Kg6=) (10. Kf8=) (10. Kg8=)
(Read Only)pid=17173
(17) Posted by Geir Sune Tallaksen Østmoe [Monday, Apr 8, 2019 14:44]

And for my position: 1.Rh8+ Sxh8 2.Rxh8+ Sf8 3.Rxf8+ Re8 4.Rxe8+ Bc8 5.Rxc8+ Kxc8 6.b7+ Qxb7 7.Bxb7+ Kxb7 8.Sxa5+ Ka6 9.gxf4 Kxa5 10.f5 b3 11.f6 b2 12.f7 b1Q 13.f8Q and now Black has many queen moves that keep the position drawn. Black’s move 3, 4 and 5 are forced, and are therefore not ‘!’ moves by the definitions above.
(Read Only)pid=17176
(18) Posted by Geir Sune Tallaksen Østmoe [Monday, Apr 8, 2019 23:57]

25 half-moves again, this time with 25 '!' moves:

(= 9+11 )

1.Kxb1 bxc2+ 2.Rxc2 dxc2+ 3.Bxc2 Bxc2+ 4.Qxc2 Sxc2 5.Kxc2 Kxh8 6.Kxc3 Kg7 7.Kc4 Kg6 8.Kxc5 Kxg5 9.Kc6 h5 10.Kb7 h4 11.Kxa7 h3 12.Kb8 h2 13.a7 and now there are promotion duals.
(Read Only)pid=17179
(19) Posted by Geir Sune Tallaksen Østmoe [Tuesday, Apr 9, 2019 00:33]; edited by Geir Sune Tallaksen Østmoe [19-04-09]

I realized I can move the bishop to a4 and thereby make room for more introductory moves. 32 half-moves now.

(= 13+11 )

1.Bb7+ Rxb7 2.Rg8+ Bxg8 3.Rxg8+ Sb8 4.Rxb8+ Kxb8 5.bxc7+ Rxc7 6.dxc7+ Bxc7 7.Bxc7+ Qxc7 8.Sxc7 Kxc7 9.Kxh1 Kxc6 10.Kg2 Kc5 11.Kg3 Kxc4 12.Kxg4 Kc3 13.h4 Kb2 14.h5 Kxa2 15.h6 Kb1 16.h7 a2
(Read Only)pid=17180
(20) Posted by Bob Baker [Tuesday, Apr 9, 2019 09:16]

Geir, I think your best change was adding the black pawn at a3. Now it is an astounding feat. I don't even mind that it makes my position seem puny. My consolation is that mine ends with 11 non-captures and yours with 4. But I suspect that record may not hold up for long.
(Read Only)pid=17184

Page: [Previous] [Next] 1 2

MatPlus.Net Forum Misc Construction Challenge