The Necromancer

Great ideas killed by ignorance will rise again. Timeless values will live forever.

[The Necromancer @StelmagM]

@FRosyrat96857 @chesscom By the way again I never used Elo as my evidence in the beginning. You mentioned it first.

[The Necromancer @StelmagM]

I said 8x8 skill doesn’t generalize enough to define you as smart or a tactical genius. I argued against it from the theory of why it is a solvable, simple in comparison to variable input/infinite space games. I also flipped all your counter examples for elo: doctors, iq test winners and contract lawyers are indeed much more smart and can apply extremely more advanced strategical thinking, because they are playing the corresponding general cases.

You attacked my “boils down” to extremely large number multiplication, which is correct for the 8x8 case. You can make a 1v1 game where each player gives a multiplication problem of two large numbers to the other and they take turns guessing the results. The one whose guesses are better wins. You could even get the same elo model with the same predictability. Is this “hard”? Yes. Does it require some sort of skill? Yes. Is it boring? Yes, very. Is it an indicator of strategic thinking and tactical genius? No.

You used words like shit and shut up which is kind of lame imo.

I am not annoyed at all. Enjoy your chess. Not conceding even a little bit of the things I said though.

[The Necromancer @StelmagM]

@TCzajka Nice.

[The Necromancer @StelmagM]

@FRosyrat96857 @chesscom What is astounding to me was that I thought chess was an honorable game of decency and good qualities. You and all the absurdly mean and hateful replies prove the opposite. It got infiltrated by high ego and angry people.

[The Necromancer @StelmagM]

Man I keep answering in good style with actually correct arguments (yes they are the similar, I just explain them every time according to your counter argument so you get it) and you keep insulting me. In the previous answer I even clearly made the argument about elo flipping all your examples (not just about O(1), which is central anyway). The point is that my arguments were correct from the beginning. I adopted a more scientific language because I thought you understood something. You never refuted any of my arguments really. Whatever go play chess. I don’t concede anything at all though. You are free to believe that your 1700 elo makes you a mastermind. But you are not.

[The Necromancer @StelmagM]

@craigweiss Your word is true.

[The Necromancer @StelmagM]

@leecronin “Ability to correctly answer arbitrary questions, solve arbitrarily complex problems and perform arbitrarily complex tasks” is the way people mean it I think. It’s not deeper than that and doesn’t have to be.

[The Necromancer @StelmagM]

@chesscom What he is saying is more that chess players larp as strategic masterminds (they do lol). Most of them couldn’t run an engineering company though. That’s really complex.

[The Necromancer @StelmagM]

@CerryBart @cafreiman No real one can actually.

[The Necromancer @StelmagM]

@lambdabetaeta The only thing left for you to do is.. PROVE IT.

[The Necromancer @StelmagM]

@ElShogoso @chesscom Enjoy playing with your grandpa! 😁

[The Necromancer @StelmagM]

@ElShogoso @chesscom True about universality. But proves my point about kind of boring and simple.

[The Necromancer @StelmagM]

@RhysSullivan https://t.co/AhgXYvXK0Z

[The Necromancer @StelmagM]

@neerajjj6785 You cannot really “skip” them though. At some abstraction level you will hit an unskipable hurdle, even with AI.

[The Necromancer @StelmagM]

@ThePrimeagen @CodewithP Kind of means that image as floats is a more descriptive interpretation format for images than image as code as floats. Makes total sense for me.

[The Necromancer @StelmagM]

@maddoxrules “If a person has overwhelmingly more money than another, then they cannot debate him” boils down to tyranny of the poor.

[The Necromancer @StelmagM]

@WanderingJoule @jonatanpallesen For me it is more like a founder (scientist) - investor (funder) relationship. This is what I am arguing for.

[The Necromancer @StelmagM]

@WanderingJoule @jonatanpallesen Oh I see your argument now sorry. Yes it is true if you see it as an employer employee contract you are absolutely right. We are simply debating the nature of the contract. I.e. a master’s thesis does not get compensated.

[The Necromancer @StelmagM]

Again everything you said here is wrong.
Lmfao your whole argument is that any bounded, finite-input problem is technically O(1) solvable because a lookup table exists somewhere in function space.
•No. I am just arguing it is STRICTLY FAR SIMPLER than any variable input, infinite space problem.
Sure, that’s a true but empty fact. That’s like saying since the winning lottery numbers exist somewhere in the finite space of all six-digit combinations, buying a ticket requires no luck. The set is bounded, the answer is ‘out there,’ therefore predicting it is trivial. Nobody would accept that logic for a lottery. You’re only accepting it for chess because it lets you dodge the point about Elo.
•Winning lottery has nothing to do with chess. It has randomness and incomplete information. Luck ties to the incomplete information, not to the complexity class of course (although again a stochastic lottery would require STRICTLY MUCH MORE luck 🤪)
◦For elo see my previous reply*

Except it proves way too much. By that same logic, IQ tests are trivial,
•They are, assuming a fixed input that repeats itself. Assuming a variable input they are not. But 8x8s input is fixed, making it isomorphic to a trivial fixed IQ test you repeat over and over again in constant space. At some point you achieve perfect IQ relative to the fixed IQ test. Again if the IQ number is the ELO and the test is the chess game, your metaphor proves my point.

diagnosing a disease from a fixed set of symptoms is trivial,
•NOT A DOCTOR but it is assuming a 1 to 1 mapping. The point is that for real diseases, the symptom set is variable (and dynamic) and the mapping is many to many (symptom sets to diseases), which yields quite a complex problem space. If you factor in new diseases, I even think it is not Turing computable.

drafting a contract is trivial
•NO EXPERIENCE WITH DRAFTING COONTRACTS but I hink this is more about the logic and not so much about the language. Anyway still if you assume fixed formatting/requirements it is trivial computationally.
, since all of these are finite bounded tasks with a theoretical lookup table sitting in the same infinite space you’re invoking.
•They don’t all have a lookup table in the general case (like chess), they only have in the instance case, where they are arguably more trivial than chess (which means nothing complexity theory wise though)

If ‘a solution exists in principle’ erases the meaningfulness of skill differences, you’ve just argued no human skill measurement means anything, not just chess. You’re not making a point about chess. You’re making a point that destroys your own framework the moment you apply it consistently. Elo works precisely because no one has your imaginary lookup table, and skill is defined by performance without it. You’ve been dressing up a vacuous truth as a devastating insight for eight replies now. Just stop
•Again completely wrong. Some counter examples are what you mentioned yourself. People try to solve these problems in THE GENERAL CASE. A doctor’s skill is measured in the general case. It’s kind of like (although still in the general case I think medicine is Turing hard.) Magnus not knowing how big the board will be and which pieces he will get BEFORE EACH GAME.

[The Necromancer @StelmagM]

“Titled players crush weaker ones regardless of what asymptotic complexity class a fixed 8x8 instance formally belongs to.”
- Not regardless, but exactly therefore. Fully predictive elo (which is a VERY simple metric), speaks against chess complexity and skill requirement, not in favor of it.

[The Necromancer @StelmagM]

I think someone else is detouring. Nobody in this thread debated generalized chess (which is not well defined, board size being the only generalization variable is intuitive but arbitrary).

“And your “single multiplication” solves it only in the nonconstructive sense that a lookup table exists somewhere in the space of all functions.”

-Again this is wrong due to constant size. We don’t need the space of all functions for the 8x8 case. Of course you cannot solve an exponential chess generalization in constant time. But that’s my point. People are not playing an abstract game, they are playing a concrete, constant one. Note that you can also generalize ANY constant size game with bound configurations and fixed input to exponential complexity. Does it mean that match 3 is as complex as chess?

“Nobody has it, nobody can compute it, no one will in the lifetime of this universe.”

-Unprovable claim due to proven solvability which is the single thing I argued in favor of (more like tried to explain, I don’t have to argue in favor of it because it is proven). Potential counter argument (that mean nothing because we are trying to prove or disprove your unprovable claim), is the exponential growth of computing capacity and LLM based computation.

“Existence in principle isn’t a solution, it’s a tautology dressed as an insight.”

-Not a solution, neither a tautology of course. Solvability (again what I refer to, not a concrete solution, even not the weak one) is basically a proof of a Hilbert epsilon in the general case. In the case of 8x8 in your generalization model it is of course trivial, which speaks about the triviality of the 8x8 and chess but doesn’t make all Hilbert epsilon claims tautologies. Of course you can say that the proof of one instance does not make the game trivial. For this note that for any arbitrary but fixed n’, the n’ x n��� case is still constant and therefore trivially solvable. In that way, generalization of chess based on board size takes you nowhere. You need to introduce a mechanism that imposes some infinite series at least. (Maybe a board that grows dynamically after a certain amount of moves (still very weak) or some kind of randomness.)

-“None of this touches the actual claim, which is that rating systems measure something real. Elo predicts outcomes with brutal consistency across millions of games. Titled players crush weaker ones regardless of what asymptotic complexity class a fixed 8x8 instance formally belongs to. You’ve built an elaborate detour around the point instead of engaging it.”

Your argument eats its own tail. As far as I know elo is not measured in a (λn. n x n) board (which actually doesn’t exist as a single physical object, it is an abstraction). I didn’t have to refer to the elo argument. Of course it measures something real and is predictive (by construction and trivially also, since fixed n chess skill is purely learnable, again because it is constant). In our case it measures skill for 8x8 chess, which is not a really complex game, and therefore this skill for me does not generalize well enough.