Battle systems in RPGs have always been a hotly debated topic. You have the "I only play command-based JRPGs" group, the "strategic turn-based isometric or death" group, the "everything but action is archaic" group and so on and so forth.
While I don't expect all of them (or even any of them really) to agree, I've thought a lot about what makes RPG battle systems tick, and why I find some of them continuously intriguing and others not so much. What it boils down to is - I believe - state space. An RPG battle system ideally needs to have a state space that is as large as or larger than the number of turns you will take throughout the game. However, I've always had some trouble explaining what this means clearly and succinctly.
I now think the answer are gambits. Not as a system, but as a tool for quickly evaluating how interesting a battle system is.
Think of a simple "programming language", like gambits. You have a number of if/else branches and statements for each character, based on properties of the ongoing battle, which decide their actions. Our metric for the state space is then the complexity (e.g. concretely the number of branches) of the "program" you need to complete 95% of the battles in a game successfully.
This unifies many points that I've experienced while trying to think of why one battle system is more continuously engaging than another, and captures their effect on the size of our battle state space:
Battles in traditional command-based JRPG systems without any additional gimmicks, no meaningful positioning or environmental effects and no meaningful resource management are incredibly easy to automate, which is one reason they are getting progressively more scarce.
While I don't expect all of them (or even any of them really) to agree, I've thought a lot about what makes RPG battle systems tick, and why I find some of them continuously intriguing and others not so much. What it boils down to is - I believe - state space. An RPG battle system ideally needs to have a state space that is as large as or larger than the number of turns you will take throughout the game. However, I've always had some trouble explaining what this means clearly and succinctly.
I now think the answer are gambits. Not as a system, but as a tool for quickly evaluating how interesting a battle system is.
Think of a simple "programming language", like gambits. You have a number of if/else branches and statements for each character, based on properties of the ongoing battle, which decide their actions. Our metric for the state space is then the complexity (e.g. concretely the number of branches) of the "program" you need to complete 95% of the battles in a game successfully.
This unifies many points that I've experienced while trying to think of why one battle system is more continuously engaging than another, and captures their effect on the size of our battle state space:
- Obviously, meaningful positioning will greatly increase the number of situations our program needs to take into account.
- Environmental effects add an entirely spearate axis to the state space yet again.
- As will effects which are situational and/or specific to some enemy type.
- If the game has meaningful resource management (like a dungeon crawler) then we need to take our resource levels into account, otherwise not.
- More distinct types of enemies with different attack patterns help, but they only increase our metric linearly (unlike e.g. positioning).
Battles in traditional command-based JRPG systems without any additional gimmicks, no meaningful positioning or environmental effects and no meaningful resource management are incredibly easy to automate, which is one reason they are getting progressively more scarce.