I’ve been wrangling over which perspective to use for the new game I have in development (the one I dropped a reference to in the Sydney Morning Herald article). I use the term ‘development’ to loosely describe the design phase of the game: a collection of reference pictures from Google images, a half-completed game design document, Photoshop layouts of all the screens (with a lot of crappy, hand-drawn images) and a massive, complicated spreadsheet containing all the game logic.
Actually, that last one I’m particularly proud of. Many designers don’t like to admit it, but Excel is an awesome tool for testing complex logical interactions in your games. It’s too easy to write ‘when the player does x, y should happen’ in a game design document but when your programmers try to interpret that in code, it simply doesn’t work. And it’s because YOU as a designer got your logic wrong. Excel eliminates that problem. You can’t fudge logic in a spreadsheet. If it doesn’t work in Excel, it won’t work in your programming language of choice. And if it does work in Excel, it makes the programmers job so much easier because it’s simpler him/her to lift logical interactions from another program than it is to interpret pseudo-code from a Word document.
But that’s not what I wish to rant about today. Today’s rant is about how to choose the graphical representation for a game. There are many excellent articles about this already (one of my faves is this one) so I won’t regurgitate the good words of other designers. But some of these articles don’t take the concept far enough. They talk about what the different types of graphical projection are, but don’t really talk about why you would choose one over the other (save for technical reasons). So here is my take on it.
Graphical projection is a lot more than a technical issue. On a phone (which is mainly what I design for), stepping up to true 3D is a massive technical and resource hurdle (both in the sense of computing resource on the phone and the time and money it takes to generate 3D assets). It really is outside the scope of most indie games, and I’m definitely not touching it. So I know I’m going to use 2D. For me it basically comes down to these 2D projections: orthographic or isometric. I’ll leave out any discussion of tiled versus fully drawn levels as I believe this is a separate consideration to graphical projection (maybe more on this next week). So in layman’s terms, it’s either top-down view, or ‘2.5D’ (viewing your character from an angle on the side).
The key question is: can you identify enough with your character in a role-playing game (RPG) if you never really see his/her face? This is the problem with the top-down view. Sure, you’ll see his/her face in the inventory screen and character customisation screen. But during the game, during the action, you’ll only see the tops of their heads. Will that provide enough immersion and empathy as a game experience? The flipside, of course, is that the amount of art that needs to be generated is 1/8th the requirement for an isometric view. For every pose in isometric view (e.g. your character just standing there) you need to draw that image 8 times. Now imagine that character swinging a sword. Imagine you want at least 3 frames per animation. That’s 8 x 3 = 24 drawings to represent one simple thing. In the top-down view, it’s 3. That’s a friggin’ massive difference. I could devote those resources to adding more frames (smoother animations) or draw more monsters or whatever. Or I could draw a different sequence for swinging a sword as opposed to thrusting a spear (one of my pet hates in RPGs: the animations are the same no matter what weapon the character is using).
Also, it’s actually easier to represent scale in a top-down world. Fighting a dragon? No problem – he’ll take up 9 tiles instead of 1. In an isometric view, you have to take into consideration the z-axis on any screen, which makes it difficult to represent very tall and normal sized enemies on the same screen. There’s a bunch more design issues I could get into, and I have’t decided 100% yet, but I’m strongly leaning towards top-down simply due to the amount of stuff I could put into the game. Here I am talking about technical issues while chastising other designers for doing the same.
Is it a deal-breaker if you don’t see your character’s face during the action in a game? Of course, I’d be interested if anyone had any thoughts or experiences of their own!