﻿﻿ MatPlus.Net

Website founded by
Milan Velimirović
in 2006

18:18 UTC
 ISC 2020

Remember me

 CHESS SOLVINGTournamentsRating lists1-Oct-2020
 B P C F

The codex states that Dead Reckoning doens't apply unless stipulated. Who would be such an idiot to ignore that for a century old problem? :)

I had an interesting thought lately about Loyd's famous Kriegspiel problem and modern day law.

Samuel Loyd, Le Sphinx 15/10/1866
(= 3+0 )

Place BK for-1) Mate, 2) Stalemate, 3) #1, \$) & A square on which it can never be mated

When looking over the 4th stipulation, I noticed that the square h3 is technically a square that the Black king can never be mated on, as it is forced to capture the queen and then it can't be mated. But we know that's not what Loyd meant.

So, I propose the following addition to Loyd's problem that meshes it with modern day-Place BK for-5) A dead position

Of course, the only possbile answer is to place the Black king on h3. In my opinion, this stroke of luck prodcues lovely harmony between past and present.

What do you think?

What is the difference between h3 and f3?

Oof-I didn't see that. Two solutions isn't worthless though,

h4 too

Okay, now my suggestion is worthless. It should be doable in some other from though.

(6) Posted by Andrew Buchanan [Friday, Aug 14, 2020 19:59]

I never thought of that, James: pity the DP solution isn't unique, but I'm sure there must be another such position where DP is unique.

The codex actually states:
QUOTE
Unless expressly stipulated, the rule of dead position does not apply to the solution of chess compositions except for retro-problems.

Twin d) (place bK on a square where it can't be mated) is cooked, whether or not DP rule applies. It has always been cooked. It could be salvaged by insisting on a living position, maybe.

The proposed DP twin e) would also be cooked, as discussed. Mentioning DP in one twin would trigger DP rule for that twin, but not for other twins, I always think.

The term "retro-problem" was deliberately not defined in the Codex, and the term also appears in the convention on the 50-move rule. I think the best practical approach to define "retro" comes from the philosopher Ludwig Wittgenstein, see: https://medium.com/@magne/what-did-wittgenstein-say-about-what-a-game-is-d333383cf8b4. The metaphor of the ship's cable is particularly beautiful.

Normally, "add piece" problems are regarded as retro because legality of the position is a core concern. But here all the vacant squares except for those adjacent to wK are legal for bK to occupy. So to call this problem "retro" is a bit lame: it's like saying that a problem with Black in check and hence Black to move is retro. Strictly one might argue that retro logic is indeed used in a negative way, but it's cringeworthy. So I'm kind of glad that this one is unsound.

How does change of rules affect a problem composed in the past, though? Generally a problem is as sound as the maximum point of its soundness in time: this is the Golden Age principle which e.g. protects dummy pawn problems which were orthodox in 1862. But this problem has always been cooked, although its only just been recognized - and the simplest correction ironically uses the notion of "living position" which is a modern one since the concept of DP was articulated in 1997!