What makes creating a fighting game tutorial difficult? I assume it's difficult because I only know of small indie games and Granblue Fantasy Versus that have been praised for a good tutorial (and also because I have not felt well-taught by other fighting games, even a party game like Smash Bros.).
Fighting games essentially have two major stages of player learning. The first is the “physical” stage where they learn to perform the physical elements of playing the game. This includes learning which button maps to which attack, what the basic controller motions are, how to throw a fireball, how to perform a command grab, and so on. The second stage is the “mental” stage where players learn more abstract concepts, like when to use specific moves and how to play your opponent instead of just performing moves. Fighting game tutorials today tend to exhibit the easy stuff - mostly teaching the physical stage rather than the mental stage.
Each of these stages has a range of elements that can be sorted from easy to difficult. The high end of the physical stage is executing precise frame-perfect input timing under pressure, like the famous Evo moment 37:
It’s obvious to the audience here that this is a difficult feat to execute. It requires extremely precise timing in order to go from what you want to do (parry every hit) to physically performing it without making a mistake. This requires practice and training. However, what most players don’t often recognize is the high end mental stage on display, like so:
“What’s so special about that? He’s just doing the uppercut over and over!”
This is actually a special situation. All the Adon player has to do here is block. Ryu’s fierce uppercut is extremely punishable if blocked. Both players (professional gamers each with thousands of hours of experience) know this. However, by this point in the match, Adon has already been so mentally flustered by the set of circumstances - a long tournament, a difficult battle through the loser’s bracket, the fact that Ryu seems to know exactly what he’s going to do before he does it - that he’s panicking. He’s too flustered to think strategically at this moment so he’s resorting to his base habits. Ryu knows this, so he can do the high-risk high-reward move (fierce uppercut) with impunity because he knows Adon is too panicked to properly defend and that doing so will continue to fluster Adon. If Adon were in a right state of mind, this strategy would never work. But because he’s mentally guard crushed in this moment, he’s vulnerable to it. This Adon desperately needs to take a minute to refocus, to get out of that bad mental state, in order to have a chance at winning. Recognizing your opponent’s mental state, what your actions are doing to affect your opponent’s mental state, and adjusting your play based on that reading is what the mental stage of fighting games is all about.
Unfortunately, fighting game tutorials don’t do a great job of teaching the mental stage elements of fighting games. Tutorials must be able to recognize and teach abstract concepts like a “mixup” - attacking with an angle and timing that your opponent cannot normally react quickly enough so they must guess how to defend it. This kind of behavior can be very difficult for an AI to emulate, and most players themselves often have a hard time recognizing such things without having earned that experience through a lot of hard-fought losses and self-reflection. I do think it is possible to do through a significant investment in game AI and the tutorial in general, but fighting games tend to be smaller budget affairs and don’t often have a lot of room in the budget for tutorials for newer players when so many of them already understand many of the basics and don’t want to sit through tutorials. I suspect that the best way to go about doing this would be to incorporate such behaviors into the single player elements of the game so that players get used to the idea of looking for such things. For example, maybe this AI tends to fall for crossups, that one likes pushing buttons on wakeup, maybe even giving them a visible scale from "calm” to “flustered” at lower difficulty levels or something.
[Join us on Discord] and/or [Support us on Patreon]
The FANTa Project is being rebooted. [What is the FANTa project?]
Got a burning question you want answered?
Short questions: Ask a Game Dev on Twitter
Long questions: Ask a Game Dev on Tumblr
Frequent Questions: The FAQ
