Hello, I'm starting, as soon as my Java is good enough to my own taste (which means either next month assuming I have the time, or this summer if I don't), a fighting game from scratch as a personal side project for school. I prefer C over Java (heck, I prefer Nasm over Java), but the point of doing this is to learn, so I guess I'll deal with it somehow. This is going to be the first video game I'll design, I've only created a couple board games so far (and imagined hundreds of unrealistically complex games that I'll never be able to code, like every beginner does haha)

Now, I've been a competitive fighting game player for the past 6 years. I know what is 'good' and what doesn't work, from a competitive point of view. But professional players are usually very attached to their games, even when skill transfers from one game to the other, and I won't be able to build a player base just from already good players. So I need to make my game as accessible as I can make it. I realize, though, that my inexperience as a game developer/designer and my lack of free time may stop me from making the game as good as I would want it to be, so the general idea I'm following is 'don't focus on the amount of content, focus on its quality'.

So if you were to make a fighting game, how would you, as a professional game designer, do it ? I'm being purposely vague here, feel free to talk about whatever you'd like, you can be sure that it is going to be relevant to me : would you start with the engine or the menu, would you start with the universe and build the gameplay upon it, or find a story to justify your gameplay ideas, or even why you think that fighting game is not a viable genre if you think so.


Just for the record, I'll sum up the most important aspects of the game I'm trying to make, but don't let it influence your answer. If anything, use it as a basis for critique and tips.

The game will be on Ouya, or whatever Android based console cool kids play when the game is ready. Its key element is that it focuses on making the player build their own character and using it in story mode, online ranked matches, and everything else, instead of using the character roster like in other fighting games. The game will have a free version for tournament play, which only consists in versus mode with the ability to save replays, and a pay to play version which will allow the player to create his character and play the other modes (story mode, online mode, etc...)

The story mode is basically a big tutorial that has enough information (which I've somehow managed to disguise into story elements) to make a casual player good enough to attend tournaments and not get humiliated. As the player goes through the story mode, he unlocks new moves, aesthetics and other collectibles for his character, which act as a candy-coated way to force players into following the tutorial.

The game still has a roster, which can be used by casuals to show very quickly and easily how the game works to their friends, and has uses for tournament play. It consists in the 8 main characters in the story mode, the 8 bosses if you will.

Customization is a rather vague term, there are many way it could be interpreted so here's my version. This is very likely to change, but so far, each character has a set amount of points he can spend by assigning moves to the list of possible inputs. The better the move, the more point it costs. The reasoning behind that is very long and boring, but to sum it up, it allows easy balancing from my part, is easier to pick up for newer players than other systems I've thought of, and puts severe limitations to the maximum potential of a player-created character, which is what I want.

To maximize the accessibility, the game does not follow the standard 6 attack button, push back to block, push up to jump setting other fighting games follow. It has 2 attack buttons, a jump button, and a guard button. Of course, I let the player decide if he wants to use pbtb and putj, but that's just an option. The 2 attack buttons behave really differently. One is a magic attack button, the other is a normal attack button. Normal attacks have a lot of variety, magic attacks have a theme (the story involves you being the master of elements and doing elemental magic) and use mana, which means they can't be spammed and allow me to have a big control over the flow of the match by tweaking the mana cost of an attack that turns out to be overpowered. Combos are limited thanks to a system I found in Smash Bros, which I will try to adapt to a 2d traditional fighting game, that will allow me to not worry too much about infinite combos when designing the different moves.

As for graphics... this is the big question I'm trying to answer right now. 2d sprites are prettier, but sprites and customization don't go along well together. 3d fighting games are simpler to implement with a ready-to-use 3d engine (and there's plenty of them, Unity has just been ported to Ouya I heard), but they're ugly when they don't have the budget Capcom has. And as it turns out, I don't have the budget Capcom has.

There's another issue I'm not yet able to solve on paper, IA. But that's more a programming issue it doesn't really belong here. I'll figure it out with time eventually.

Hi, this is going to be my first post here (after a few discarded drafts where I decided I was too inexperienced as a game developer to publish, I must say), glad to meet you guys !

So I'm just starting off game designing as a hobby, mainly as a way to try all the cool stuff I learn in college, where I follow maths and IT science courses.

But all the ideas I seem to be getting seem to come up in the same situations : I'm playing a game, and I see something wrong in that game. A design flaw, a bug, or whatever that can be 'wrong'. Then I try to work a solution, without even thinking about if it would fit the game that had this problem in the first place or not, and I build a completely different game around that solution.
As an example, last time I played Skyrim, I found myself save-scumming a lot, which completely screwed up the immersion. So I created a story about a time traveling spy who uses his powers (aka save-scums a whole lot) to solve puzzles an achieve great stuff. I went from the assumption that save scumming breaks the immersion, to a game where save scumming is the main mechanic, and actually helps the immersion.

And I feel like I can't really get new ideas with other methods, if that makes any sense. So here's my question to you : do you also stick to some kind of method when you find ideas, or is your work only driven by hazards of creativity ?