Website founded by
Milan Velimirović
in 2006

2:12 UTC
ISC 2020



Remember me

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

Rating lists


MatPlus.Net Forum General An entertaining tale about Dead Reckoning
You can only view this page!
Page: [Previous] [Next] 1 2
(1) Posted by Steffen Slumstrup Nielsen [Wednesday, Nov 27, 2013 16:28]

An entertaining tale about Dead Reckoning

I came across this blogpost. I don't know if the examples are original, but the story certainly is.
(Read Only)pid=11177
(2) Posted by Andrew Buchanan [Saturday, Nov 30, 2013 12:48]

Thanks for the link to this fascinating article. Its subject, I think, is not dead reckoning (= "inevitable draw"), but on DR's trickier cousin, "inevitable non-loss". This only applies in penalty situations: flag-fall (6.9) 3rd illegal move (7.4.b) and mobile phone sound (12.3.b). In the first two cases, under the current 2009 FIDE rules, the game is drawn, if the position is such that the opponent cannot checkmate the player’s king by any possible series of legal moves. In the third case (mobile phone violation), the penalized player will always lose, but if the opponent cannot win the game by any series of legal moves, his score shall be a draw.

So the most practical case of planning ahead is where one player is likely to lose on time, but is ahead on the board. In this case the battle becomes whether he can close off all possible lines of checkmate for his opponent, by forcing a position of insufficient material FOR THAT OPPONENT, or stalemate, or blocked position.

However, with respect I don't think that the logic in this fine article quite works. I have three issues. The first is quite messy, but the most important, unfortunately.

With his hands flying nervously all over the board, T knocked his own king over, picked it up and managed to put it on d8, but the clock was already flashing with a row of zeros on Black's side.
"I seem to have lost on time", he confessed, but all of a sudden recalled what he had learnt in the weekend.
"No, wait, it's a draw because of insufficient material!"

The relevant rules I think are these:
6.2.a. When using a chess clock, each player must make a minimum number of moves or all moves in an allotted period of time and/or may be allocated an additional amount of time with each move. All these must be specified in advance.
6.3 Immediately after a flag falls, the requirements of article 6.2 a. must be checked.
6.7.a. During the game each player, having made his move on the chessboard, shall stop his own clock and start his opponent’s clock. A player must always be allowed to stop his clock. His move is not considered to have been completed until he has done so, unless the move that was made ends the game. (See the Articles 5.1.a, 5.2.a, 5.2.b, 5.2.c and 9.6) [checkmate, stalemate, dead position, agreed draw, and (redundantly) dead position again!]
6.8 A flag is considered to have fallen when the arbiter observes the fact or when either player has made a valid claim to that effect.
6.9 Except where one of the Articles: 5.1.a, 5.1.b, 5.2.a, 5.2.b, 5.2.c applies, [effectively the same list of criteria as before, together with resignation.] if a player does not complete the prescribed number of moves in the allotted time, the game is lost by the player. However, the game is drawn, if the position is such that the opponent cannot checkmate the player’s king by any possible series of legal moves.

So what's the sequence of events? T plays the move, then makes the claim of flag fall. So by 6.8, the move happened before the fall. Therefore by 6.3 we check the number of moves under 6.2.a. Under 6.7.a, the move is not complete (for timing purposes) unless it ended the game. But it did end the game, so under 6.9.a, the game ended in dead position, by 5.2.b. The point is that the final sentence of 6.9 about inevitable non-loss never gets to apply.

If flag fall had been claimed before the king was placed on d8, as the text suggests it could have been, then the requirements of 6.2.a must be immediately checked, i.e. before the king gets to move. Prior to 66...Kd8, the position can end in a mate for either player, so there is no escape for T. Inevitable non-loss doesn't get to apply.

Even if we have a photo finish, where the flag fall claim happened *as* Kd8 is being played, we still don't get to trigger the inevitable non-loss clause. Because when we come to evaluate the position, the arbiter must surely decide either it's going to have Kd7 (alive for both players) or Kd8 (dead).

For centuries, problemists and study composers have thought that the only motives for promotion to rook or bishop are forcing or avoiding stalemate. But tonight our friend B has demonstrated, with his extraordinary 66th move, that there is another motive: retaining sufficient material.
In the final position, with White to move after Black overstepping time, White has sufficient material because on the forced continuation 67. Rxc8+, Black has the legal move 67. - Kd7 after which he can in principle be checkmated. Had the rook on b8 been a queen, the only legal continuation would have been 67. Qxc8+ Kxc8 stalemate, and therefore 66. - Kd8 would have ended the game.

Alas no: it's about avoiding stalemate, not insufficient material. White will always retain Pa7 - the problem is that after 67... Kxc8 White is stalemated.

The task was unusual: Draw in 1 move.

OK, so there is no time pressure here, and hence "inevitable non-loss" is irrelevant. After any White move, 1.b8=X, there is always the line 1... Qxb8+ 2. axb8=Q+ winning, so there is no way that the game can end in a dead position after White's next move. And clearly there is no stalemate after White's move. Certainly if *White* is under time pressure, it makes sense to promote to B rather than Q, because that locks in the capture of BQ the following turn.

I have steered clear of inevitable non-loss as a source of compositions, because I thought it was too tricksy for my simple mind. I preferred to stick with with the much simpler area of inevitable draw. But good for someone to brave these waters. I am genuinely sorry to raise these points. I think the article is delightful, and I hope the author manages to fix it. Or maybe the laws were significantly different 10 years ago, in Linköping, at the time of the original incident? I haven't the appetite to check that.

I would post something like this as a reply to the article, if I can figure out the Swedish input buttons.
(Read Only)pid=11178
(3) Posted by Andrew Buchanan [Saturday, Nov 30, 2013 16:53]

(= 3+4 )

White to play.

If Black runs out of time just after moving to this position, but before hitting the clock, he can relax. This is not a dead position, but White can never win. So the penalty insurance clause is triggered, and the game is drawn.
(Read Only)pid=11179
(4) Posted by Geir Sune Tallaksen Østmoe [Saturday, Nov 30, 2013 23:15]

Thank you for the entertaining article!

Regarding Andrew's reply, I agree that from s problemist's view, the examples are not perfect. The way I read the article, I think its subject is neither dead reckoning nor inevitable non-loss, but insufficient material. Correct me if I am wrong, but as an OTB player, I have always thought that insufficient material was a term meant to describe that one player has no way to win by checkmate. Thus, both dead reckoning and inevitable non-loss are cases of insufficient material. For a problemist, insufficient material is therefore a vague term, while for an OTB player, it is the relevant term to consider when one player loses on time.

If this view is correct, Andrew's points 1 and 2 are irrelevant for the article's subject, but it would surely be nice if someone found a way to let the examples describe inevitable non-loss.

Regarding point 3, I interpret "draw in 1 move" as a variant of the stipulation "draw." As you surely know, the stipulation "draw" means that White has to achieve AT LEAST a draw, while some of Black's replies might lead to a win for White instead. So I think the stipulation makes sense, but maybe "inevitable non-loss in 1 move" would have been a clearer stipulation?
(Read Only)pid=11180
(5) Posted by Hauke Reddmann [Sunday, Dec 1, 2013 15:13]

Clearly. Although "inevitably non-loss" is a word monster -
the fairy dudes should coin a new word for it. In honor of
a certain depressive bread loaf with much too short arms,
I provisorily term it "gnampf" for this post,
and "dead reckoning" is then a reciprocal gnampf situation.

(Read Only)pid=11183
(6) Posted by Andrew Buchanan [Monday, Dec 2, 2013 02:09]

Geir, Hauke,

Thanks for the thoughtful replies.

None of the words "insufficient", "mating" or "material" occur in the current Laws of Chess. A quest scout around the internet suggests that there are (at least) two senses of the phrase "Insufficient Mating Material", one broader than the other. E.g. compare with I had thought there was only one sense, the narrower one. (Position K in the first link is quite fun, by the way.)

As for the third point, maybe we are intended to figure out how to interpret "Draw in 1 move"? This is not an unfair ask by the composer, since the composition is so basic, and so much context has been given by the surrounding story.

(Read Only)pid=11186
(7) Posted by Kevin Begley [Tuesday, Dec 3, 2013 03:16]; edited by Kevin Begley [13-12-03]

>"Although "inevitably non-loss" is a word monster - the fairy dudes should coin a new word for it."

OK, how about.. we start with terms for the players who move in dead positions:

Zombie: position is dead, though it continues moving -- send more brains!
"In your head, in your head they are fighting..." -Cranberries (Zombie).

That suggests two options...
1) (Un)Dead Reckoning, and 2) ZomBite: position isn't yet dead, but one player was bitten.

Incidentally, I think the rules of chess should be modified, such that "Dead" is not a RULE OF MOVEMENT, but instead a CONTINGENT RULE (which changes according to the stipulated objective).
And, the onus of correction falls especially heavily upon retro/fairy problemists...

In FIDE-Chess, the stipulated objective is always implied: checkmate.
As a result, dead positions are defined according to this implied objective; however, this offers no contingency for alternative objectives (which may be stipulated in a fairy chess problem, or chess variant game).

The definition of dead position must be based upon the possibility of a "help-objective."
In FIDE-Chess, the rules of movement depend upon the possibility of help-checkmate.

But, consider the famous queen-odds game, where one player is required to selfmate (I think most every player knows that game -- and, the curious player will ponder how this affects the termination states of the game).
The rules of movement should be defined according to the possibility of each player's help-objective (in this case, help-selfmate).

In order to properly define alternate games/conditions -- such that they "default to the rules of (FIDE-)Chess" -- the rules of movement must be drafted to facilitate reasonable contingencies (including alteration of an implied objective).

It's not only for the benefit of fairy problems and variant gamers -- it also helps clarify the rules, for orthodox chess players.
Even if nothing else, it would allow players to more easily interpret termination states, when presented with an alternative challenge.

In anticipation of the dreaded -- and dreadfully common -- counter argument ("we will never convince an orthodox chess federation to accept even the most reasonable contingency rules"), I would continue to urge open minded chess enthusiasts to rage against the instability of this passive fortress position.
1) There is no value in a nihilistic resignation to the false conclusion of inevitable failure (problemists have a long history of helping push FIDE into better rules), and
2) In the unlikely event that FIDE should adamantly refuse adaptation, then fairy problemists would be called to actively draft a separate rule book (even if it means that we stand apart from FIDE -- a false affiliation, in the interest of title-affiliation, is no affiliation at all).

In either event, the choice is individually ours: adapt, or ... ?
(Read Only)pid=11189
(8) Posted by Per Olin [Wednesday, Dec 4, 2013 11:34]

It is surprising how neglected the dead reckoning is among endgame study composers. Regularly one can in EG see studies with the given solutions ending in stalemate, when actually they end earlier in a dead position.

Latest example of this is from the 3rd FIDE World Cup 2013, preliminary award The second prize winner is claimed to end in an ideal stalemate with three pins. Unnoticed is that with the queen promotion the play ends immediately without ideal stalemate according to the FIDE rules of chess to which the WFCC Codex of Chess Composition refers.
(Read Only)pid=11193
(9) Posted by Valery Liskovets [Wednesday, Dec 4, 2013 13:34]

Kevin has touched the similar idea wrt other stipulations including checkmate.

In fact for checkmates, no special additional regulation in the Laws of Chess is necessary. It suffices to introduce the appropriate new stipulation that generalizes naturally the ordinary checkmate. My article "Completely unavoidable mate" just published in The Problemist is devoted to this idea. Here is one of its diverse examples, where ###1 denotes (direct) "completely unavoidable mate in one move":

VL, The Problemist, Nov.2013
(= 6+4 )

1.c7### (since further unavoidably 1... xc7+ 2.xc7#/dxc7# -- this proof-play lies outside the intended solution).

Valery Liskovets
(Read Only)pid=11194
(10) Posted by Hauke Reddmann [Wednesday, Dec 4, 2013 13:41]

A self-mate, so to say :-)
(White forces Black to force White to mate.)
(Or: White forces Black to s#1 himself.)
Now, could Black force White to force Black to force White to mate? :-)

(Read Only)pid=11195
(11) Posted by Geir Sune Tallaksen Østmoe [Wednesday, Dec 4, 2013 20:05]; edited by Geir Sune Tallaksen Østmoe [13-12-04]

Something like this?

(= 4+6 )

(Read Only)pid=11196
(12) Posted by seetharaman kalyan [Wednesday, Dec 4, 2013 20:34]

This position is not unavoidable selfmate. White has the alternative 1.d2.
(Read Only)pid=11197
(13) Posted by Sven Hendrik Lossin [Wednesday, Dec 4, 2013 20:57]

Nice one.
The intended solution probably 1.b3 s###.
But as far as I understood Mr Liskovets it is s###1 because one move is enough to ensure the unique result.
(Read Only)pid=11198
(14) Posted by Geir Sune Tallaksen Østmoe [Wednesday, Dec 4, 2013 23:23]

Hmm, yes, you are probably right, Sven. When I read the definition the first time, I thought the number referred to the number of moves to mate, but reading it over again, I agree that my problem is S###1.
(Read Only)pid=11199
(15) Posted by Valery Liskovets [Thursday, Dec 5, 2013 10:32]

Very well! s###1 indeed, in the "no-brainer" style. This is the very first CU-selfmate (more precisely, self-CU-mate) problem (CU = completely unavoidable). A pity, the last W move is not exact.
(Read Only)pid=11200
(16) Posted by Kevin Begley [Thursday, Dec 5, 2013 16:27]

Wonderful thread, this!
(Read Only)pid=11201
(17) Posted by Andrew Buchanan [Monday, Dec 9, 2013 18:00]; edited by Andrew Buchanan [13-12-09]

Wow so much to respond to!

OK first thing to remark on is the ### notation. I like it! That would give us === for dead. (Curious there was never a notation for that.) And maybe something like >>> for non-loss?

By coincidence, the latest "Arbiter's Notebook" column in chess cafe includes a position by myself which is already inevitable mate. One can retract WBd5 to some other square on the a2-g8 diagonal to leave an s###1 position.

I don't mind if traditionalists compose problems or studies that would prematurely run into death. Although the Laws & Codex are clear about dead positions, the most important sentence in the Codex is:

It is not intended to be a body of established law which problemists must observe on pain of being condemned of heresy or worse; problemists are independent spirits, and it would be pointless for the PCCC to attempt to legislate in that way.

It would be a shame to lose any wonderful composition just because of A1.3, which hasn't been around for ever. We can have the best of both worlds. As long as no one tries to pretend that DR is unorthodox, I am OK! :D

On the other hand, there is an opportunity to make new stalemate problems and studies to rely on deadness for soundness. I have a couple of economical h=3 somewhere on my website which I should dig out. I should try to push them towards a sympathetic magazine.

I like the 1.Rc7### problem. I think multiple mating lines in the "proof-play" do not constitute defect in the slightest. The situation is just like Black's mating move in a self-mate - indeed arguably the more lines the better.
(Read Only)pid=11218
(18) Posted by Neal Turner [Tuesday, Dec 10, 2013 12:48]

Am I the only one who supports Mr Petkov's view that mating duals in a selfmate are in fact a defect?
(Read Only)pid=11219
(19) Posted by Sven Hendrik Lossin [Tuesday, Dec 10, 2013 14:52]

Mating duals in selfmates? What is that?
To me "duals" only affect the side that is responsible to fulfill the stipulation so that "dual" is a wrong term when it comes to black play in a selfmate.

I have seen some judgement that respected those kind of "duals" or however you might call it and I think it is mostly misjudgement.
There are cases when multiple black moves detract for example:
In a s#2 there are three variations with
In that case it obviously lessens the thematic value but I think it is wrong to see multiple black mating moves as a defect generally.
(Read Only)pid=11221
(20) Posted by Nikola Predrag [Tuesday, Dec 10, 2013 15:25]

If Black has more than one 1st move, that's also a defect. Since all defences will end with mate to wK, there's no point to have more than one variation (anything else should be considered as dualistic).
Actually, no variations at all, that would be a perfection. Only the diagram with wK already mated - that would be a true selfmate. And the aesthetics would depend on the artist who has painted the diagram.
(Read Only)pid=11223

Page: [Previous] [Next] 1 2

MatPlus.Net Forum General An entertaining tale about Dead Reckoning