Magic: The Gathering is Turing Complete
Determining the outcome of a Magic: The Gathering game with forced moves is undecidable.
2019/03/24 by Alex Churchill, Stella Biderman, Churchill, Alex +3 · 34 voices · 5 citations
#cs.AI #cs.CC #cs.LO
paper · pdf · doi:10.48550/arxiv.1904.09828
Abstract
Magic: The Gathering is a popular and famously complicated trading card game about magical combat. In this paper we show that optimal play in real-world Magic is at least as hard as the Halting Problem, solving a problem that has been open for a decade. To do this, we present a methodology for embedding an arbitrary Turing machine into a game of Magic such that the first player is guaranteed to win the game if and only if the Turing machine halts. Our result applies to how real Magic is played, can be achieved using standard-size tournament-legal decks, and does not rely on stochasticity or hidden information. Our result is also highly unusual in that all moves of both players are forced in the construction. This shows that even recognising who will win a game in which neither player has a non-trivial decision to make for the rest of the game is undecidable. We conclude with a discussion of the implications for a unified computational theory of games and remarks about the playability of such a board in a tournament setting.
Summary
The paper demonstrates that a universal Turing machine can be embedded into a two-player game of Magic: The Gathering using tournament-legal decks. This proves that determining the winner of a game where all remaining moves are forced is non-computable, answering an open problem in algorithmic game theory.
machine-generated · gemma4:31b
In simple words
Imagine a card game where you can use special cards to make other cards do exactly what a computer does. The writers found a way to set up the game so that one person wins only if a certain computer program finishes its work. Because some computer programs never finish, it is impossible to know for sure who will win just by looking at the game board.
machine-generated · gemma4:31b
Outline
- Introduction — Introduces Magic: The Gathering and the goal of proving its Turing completeness.
- Preliminaries — Discusses constraints on number specification to ensure transition-computability and reviews previous attempts at simulating Turing machines in Magic.
- An Overview of the Construction — Describes the high-level design of the tape, controller, and read/write head using specific card interactions.
- The Full Construction — Provides a detailed step-by-step walkthrough of how the Turing machine operates within the game rules.
- Discussion — Explores the implications for computational theories of games and the practical possibility of playing this in a tournament.
- Conclusion — Summarizes the findings and their impact on game complexity theory.
machine-generated · gemma4:31b
Argument
- A universal Turing machine (UTM) can be simulated using Magic: The Gathering cards.
The authors provide a detailed construction using creature tokens for the tape, modified Rotlung Reanimators for logic, and specific spells to force movement and reading. - The game's outcome (Alice winning) can be tied directly to whether the simulated UTM halts.
The construction uses Coalition Victory as a win condition that is only met if the halt symbol is read, triggering the creation of a blue creature. - Determining if a Turing machine halts is undecidable (the Halting Problem).
Citation of standard computational theory. - Therefore, determining the outcome of a Magic: The Gathering game is undecidable.
Logical deduction from the embedding of the UTM and the tie to the Halting Problem.
machine-generated · gemma4:31b
Assumptions
- Numbers specified by players must be expressed in standard binary notation. [stated]
- The function mapping a board state and move to the next state is computable. [stated]
- Players are using tournament-legal decks (Legacy format). [stated]
machine-generated · gemma4:31b
Claims
- Determining the outcome of a game of Magic: The Gathering in which all remaining moves are forced is undecidable. [proof]
- The equivalence of two strategies for playing Magic is undecidable. [proof]
- Optimal play in Magic: The Gathering is at least as hard as the Halting Problem. [proof]
machine-generated · gemma4:31b
Proof sketch
- The authors embed a universal Turing machine (UTM) into a game of Magic: The Gathering.
- They use creature tokens with specific power/toughness to represent the tape and creature types to represent symbols.
- Modified 'Rotlung Reanimator' cards act as the controller, triggering based on which token dies to write new symbols and move the head.
- The game is configured so that all moves are forced, making the outcome dependent solely on the the UTM's halting behavior.
- Since determining if a UTM halts is undecidable (the Halting Problem), determining the winner of this Magic game is also undecidable.
machine-generated · gemma4:31b
Open questions
- Does there exist a generalisation of Constraint Logic that explains the computational complexity of Magic: The Gathering?
It would provide a formal framework to model games with unbounded memory, which current sub-Turing models cannot. - Is the function for checking the legality of a move in Magic computable?
This is necessary to fully define the transition-computability of the game.
machine-generated · gemma4:31b
Cited by
Discussions
- Magic: The Gathering is Turing Complete [hn, 192 points, 192 comments]
- Magic: The Gathering Is Turing Complete (2019) [hn, 135 points, 70 comments]
- 4- Magic The Gathering é Turing complete. Ou seja, em teoria consegue fazer computadores e resolver problemas tipo contas ou algoritmos. Tipo, provaram isso em um paper acadêmico e tem uns vídeos diss [bsky, 20 points, 1 comments]
- oh it's better than that. it turns out that m:tg is turing-complete arxiv.org/abs/1904.09828 [bsky, 15 points, 1 comments]
- Magic: The Gathering is Turing Complete (2019) [hn, 11 points, 1 comments]
- MTG is Turing Complete. You can't say the same thing about Chess. [bsky, 9 points, 2 comments]
- @ramiismail.com brb, convincing the techbros that Magic: the Gathering is alive https://arxiv.org/abs/1904.09828 [bsky, 7 points, 0 comments]
- dá pra tentar fazer Magic rodar doom se vcs quiserem arxiv.org/abs/1904.09828 [bsky, 7 points, 1 comments]
- I don't Turing but I do Magic and thus I have been hoodwinked into putting this on my reading list. (Prompting to post provided by @anids.bsky.social) arxiv.org/abs/1904.09828 [bsky, 5 points, 1 comments]
- Magic: The Gathering Is Turing Complete [hn, 4 points, 0 comments]
- Magic: The Gathering Is Turing Complete [hn, 3 points, 1 comments]
- Magic: The Gathering Is Turing Complete [hn, 3 points, 0 comments]
- Magic: The Gathering Is Turing Complete [hn, 3 points, 0 comments]
- arxiv.org/pdf/1904.09828 [bsky, 3 points, 0 comments]
- They're doing it in my office now so I've started reading this again: arxiv.org/abs/1904.09828 [bsky, 3 points, 2 comments]
- Yes! arxiv.org/abs/1904.09828 [bsky, 2 points, 0 comments]
- Magic: The Gathering Is Turing Complete [hn, 2 points, 0 comments]
- Magic: The Gathering Is Turing Complete [hn, 2 points, 1 comments]
- Magic: The Gathering Is Turing Complete [hn, 2 points, 0 comments]
- arxiv.org/abs/1904.09828 link to the paper + how the deck works. Mindblowing stuff [bsky, 2 points, 1 comments]
- Magic: The Gathering Is Turing Complete [hn, 1 points, 0 comments]
- arxiv.org/abs/1904.09828 [bsky, 1 points, 1 comments]
- Did they just copy these guys? arxiv.org/abs/1904.09828 [bsky, 1 points, 1 comments]
- Je n'ai pas encore eu le courage d'aller vraiment lire le papier mais je crois qu'on est à peu près sûr que la réponse est non arxiv.org/abs/1904.09828 [bsky, 1 points, 0 comments]
- shades of the researchers who constructed a legacy tournament-legal turing machine in magic the gathering arxiv.org/abs/1904.09828 [bsky, 1 points, 0 comments]
- Magic: The Gathering as a algorithmic game theory Turing problem. "Even recognising who will win a game in which neither player has a non-trivial decision to make for the rest of the game is undecida [bsky, 1 points, 0 comments]
- Dans le même style : Saviez-vous que le jeu de cartes "Magic: The Gathering" permettait de réaliser n'importe quel calcul qu'un ordinateur peut faire ? Alors oui, en pratique ce serait stupidement lon [bsky, 1 points, 0 comments]
- yeah—as i said in another reply, computers mathematically *can’t* model out a game of magic, when they’ve been excellent at chess for decades arxiv.org/abs/1904.09828 [bsky, 0 points, 0 comments]
- Read the original work here: arxiv.org/abs/1904.09828 Find the decklist (slightly updated by me) here: moxfield.com/decks/38dRxW67… See an introduction to the deck by Kyle Hill here: https://youtu.be/ [bsky, 0 points, 0 comments]
- I don't think magic is precisely quantifiable actually: arxiv.org/abs/1904.09828 [bsky, 0 points, 2 comments]
- Them: "What are you doing?" Me: *taps a couple lands* "Writing code." [bsky, 0 points, 0 comments]
- a more understandable and sweet result: arxiv.org/abs/1904.09828 you can do programme synthesis with Magic The Gathering, yo [bsky, 0 points, 0 comments]
- Just learned that Magic the Gathering is technically, at least as of the publication of this paper, Turing complete and can calculate the result of any problem that is solvable in finite time. https:/ [bsky, 0 points, 0 comments]
- なんか流れてきたんだけど、これは何? Alex Churchill, Stella Biderman, Austin Herrick "Magic: The Gathering is Turing Complete" arxiv.org/pdf/1904.09828 [bsky, 0 points, 0 comments]
Related