I hit a low this weekend, feeling exhausted and dispirited (is that a word?). Finishing a game is exactly what you’d expect. A b**ch to get done with, never mind fixing bugs.
The fact is, the later stages in my game didn’t receive the attention they deserve. Now they do, and that’s good.
I’m putting together a little advice about game graphics. Maybe little use to you if your budget exceeds 10 000$. Otherwise we’re probably on the same boat.
What makes game graphics look rich?
Our perception of game graphics seems largely based on three simple dimensions:
- usability. Graphics are part of a game’s user interface. Poor ergonomics when producing graphics can mean any number of annoyances that can ruin the fun of playing, including, but non limited to:
- user can’t differentiate walls from the ground
- user can’t find their way our 3D world
- user can’t distinguish items from decorations
- user can’t see gaps between platforms
- …
- countability. A detail isn’t a detail if it can’t be counted. Twisting a curve this way that way isn’t terribly useful. Adding a couple of spikes on a flat landscape is.
- nameability. Anything that a player can name ‘counts double’. A scene full of details is ‘just full of details’. A village with a horsecar, a well and a church has 4 countable elements. If the well has a roof, you can make it 5. Tiles on the roof? 6.
The bottom line is that being a great artist isn’t necessary, and might in fact work against producing rich game graphics. Graphics for gaming is craft first, art second.
The above rules don’t invalidate whatever we hold to be principles of design. In fact I would rather assume that basic design principles (this blog’s not the place) have been assimilated before considering the above.
Drawing on a piece of paper, I like to express an idea quickly, with as few strokes as I can. But I don’t really think a game scene ever has enough details. Even if you can afford throwing in enough details to crowd a scene, getting everything in (and visually tuning your stuff to avoid the crowded look) is just what you want to do.
What will cost you dearly?
Before I started making graphics for a game, I mostly cared about organic modeling. Fair and square, I concentrated on trying to make human faces. I still don’t think it’s a bad idea and, in fact I would maintain that’s exactly the first thing to worry about. However…
- Making game graphics that interact nicely with a game engine, especially an immature game engine, can be difficult or just frustrating. Making an engine that can support just whatever graphics are brought in can be… …a nightmare. The best approach is to keep it simple (consider squares for a start… don’t underestimate all that can be done with just flat, square terrains).
- Every type of object you might want to model brings it’s own challenges. If you’re new to this, then like me you’ll have something surprising to learn about why natural landscapes are hard to make; why even square buildings with square doors and window frames can be a nag… …and so forth.
What’s not a problem (anymore)
Face and vertex count. I don’t know about textured games. All I know is that the amount of faces and vertices you throw at an iPod touch 8gb isn’t a major issue. Overlaps are an issue. So you can make a smooth, detailed model, but not stack huge faces behind each other. In 2010, even mobile 3D needn’t be rough and blocky.
Save your time and your buck?
Given the above, here are two golden rules that can help producing rich graphics in reasonable time:
*Use simple shapes*
Literally. Use cubes. ball, cylinders, circles. Simple extrusions. It’s not just saving you time; it’s making graphics easier to read for your players.
One argument against simple shapes is that simple shapes don’t amount to anything recognizable. Well they surely do. A plane means nothing. A ball means nothing. A ball above a plane means a whole world, viz. the sun, earth and sky.
*Don’t work against your 3D tool*
As a 3D artist, my 3D tool doesn’t interest me very much. Especially not the features. Every other feature seems to be doing some other stupid thing that doesn’t help creating great models.
As a game creator, I love my 3D tool. Every little feature and function can be used to achieve a slightly different feel. So instead of just using simple shapes, or spending hours fighting with meshes and bezier curves, I can make my graphics more fun and engaging in a few clicks.
While my iPad is waiting to get a programmer’s attention, I ordered a Diga pen to try out drawing software. I doubt whether that’s gonna be much use but… …Just using this thing as a testing device or a reader would be a little limiting, right?
