I’ve been working lately on implementing on-chain gaming in Chialisp. Since on-chain costs and latency are very high this requires optimization at all levels: Only two players are supported. Multiplayer creates problems in all aspects of implementation and practice.
Some of these constraints kind of sound like they'd work with Marvel Snap-- there are 6 rounds, 3 lanes to place cards in, each player has their own deck, and cards are revealed at the same time each round. There's a betting system and if you seem like you're going to lose, you're incentivized to just leave. The other rules (how many cards you can play each round, which cards are in your hand, what happens when a card is played in a certain position) are enforceable by your client.
It could be a fun example of how you can have some interesting effects built on top of a poker-like card game. The lanes usually have random effects that interact with whatever card is placed there, but the actual decisions and reveals that you're making each turn keeps the gameplay interactions between the players simple.
Some of these constraints kind of sound like they'd work with Marvel Snap-- there are 6 rounds, 3 lanes to place cards in, each player has their own deck, and cards are revealed at the same time each round. There's a betting system and if you seem like you're going to lose, you're incentivized to just leave. The other rules (how many cards you can play each round, which cards are in your hand, what happens when a card is played in a certain position) are enforceable by your client.
It could be a fun example of how you can have some interesting effects built on top of a poker-like card game. The lanes usually have random effects that interact with whatever card is placed there, but the actual decisions and reveals that you're making each turn keeps the gameplay interactions between the players simple.