At this years global game jam I did something I haven’t done in way, way too long. I animated. Sure I only animated a hand, and sure it really wasn’t very detailed, nuanced, or “good”, but when you only have a couple days to create an entire game, you do what you can.
I harbour a perpetual displeasure aimed directly at Blender. It’s not fair really. Blender is one of the most astoundingly capable open source projects ever developed, likely eclipsed only by Linux or Apache. As far as my personal evaluation goes, Blenders only real downside is that it is not Maya. Again, not really fair. Still, not being Maya has stopped me from using Blender as much as I probably should. It took me until this game jam to attempt wrapping my head around rigging in Blender.
I have created a couple of rigs in Blender previous to this jam, but they were all test rigs and simple tutorial stuff. Nothing that I would actually want to animate with. Nothing I would be confident exporting to another system. A game for example.
By the way, going through some of the tutorials for blender rigging after getting a mental grip on how it works, there are some folks out there who really don’t know what they are doing and their “tutorials” are a psychological speed bump for anyone eager to learn. I’m rarely ever confident enough in what I know about a topic to put together a tutorial, and these people illustrate why I’m probably not wrong about that. Can we draft an accord between instructors and students that states:
“This is not the definitive way to do this thing that I am showing you. This is an example of the broken way that I do things, and you would do well to seek your own way. Please ignore every instance where I emphatically state my expertise on this topic. The truth is no one knows, and anyone who claims to should be thrice examined with a wary eye.”
This time though I had picked up a new plug in for Unity called Skele. This meant that I could rig and bind a model in Blender, but I could animate in Unity. Unity shares a lot of UI elements and workflow concepts from Maya, so animating directly in the game authoring tools seemed friendlier. The real benefit was, since I was doing all the animations right in the Unity environment, the animations files pumped out were instantly available to the running game, being developed by another person on a different computer. Animation files, when handled properly, are really just glorified text files. Lists of joints and their rotations at certain time intervals. The game engine takes care of displaying the resulting character movements and deformations. The files Skele puts out could easily be put into a file repository somewhere on the internet and distributed to a team working from pretty much anywhere, without the need for the animator to be there making sure that exported file formats like fbx and collada are being imported and interpreted properly.
Game jams are pretty much failure central. At least one thing you come up with on the first day will fail spectacularly enough that it, or any number of other things, will have to be cut from the game. That’s just the way it is. You only have 48 hours to slap all this jazz together, and you may want to eat and sleep sometime in there as well. You will try something, it will fail, and you will ditch it. The upside is, you might as well try anything, since who knows, maybe that will be the thing that actually makes it into the game. Once in a while something you think might fail, works so well, and implements so smoothly, that you will wonder why you haven’t been doing it all along. A few years ago I built an entire enemy AI system using Playmaker, a visual state machine editor, and it could not have gone smoother. I think I commented at the time that the AI was about the level of an incredibly stupid roomba. The fact that I had never attempted anything like that before, but it got in the game, and it worked consistently, made me reconsider how I use Playmaker. Going forward I will be doing quite a lot of animating with Skele. Next time I’ll try to make it something more interesting than a hypnotically undulating hand.
I harbour a perpetual displeasure aimed directly at Blender. It’s not fair really. Blender is one of the most astoundingly capable open source projects ever developed, likely eclipsed only by Linux or Apache. As far as my personal evaluation goes, Blenders only real downside is that it is not Maya. Again, not really fair. Still, not being Maya has stopped me from using Blender as much as I probably should. It took me until this game jam to attempt wrapping my head around rigging in Blender.
I have created a couple of rigs in Blender previous to this jam, but they were all test rigs and simple tutorial stuff. Nothing that I would actually want to animate with. Nothing I would be confident exporting to another system. A game for example.
By the way, going through some of the tutorials for blender rigging after getting a mental grip on how it works, there are some folks out there who really don’t know what they are doing and their “tutorials” are a psychological speed bump for anyone eager to learn. I’m rarely ever confident enough in what I know about a topic to put together a tutorial, and these people illustrate why I’m probably not wrong about that. Can we draft an accord between instructors and students that states:
“This is not the definitive way to do this thing that I am showing you. This is an example of the broken way that I do things, and you would do well to seek your own way. Please ignore every instance where I emphatically state my expertise on this topic. The truth is no one knows, and anyone who claims to should be thrice examined with a wary eye.”
This time though I had picked up a new plug in for Unity called Skele. This meant that I could rig and bind a model in Blender, but I could animate in Unity. Unity shares a lot of UI elements and workflow concepts from Maya, so animating directly in the game authoring tools seemed friendlier. The real benefit was, since I was doing all the animations right in the Unity environment, the animations files pumped out were instantly available to the running game, being developed by another person on a different computer. Animation files, when handled properly, are really just glorified text files. Lists of joints and their rotations at certain time intervals. The game engine takes care of displaying the resulting character movements and deformations. The files Skele puts out could easily be put into a file repository somewhere on the internet and distributed to a team working from pretty much anywhere, without the need for the animator to be there making sure that exported file formats like fbx and collada are being imported and interpreted properly.
Game jams are pretty much failure central. At least one thing you come up with on the first day will fail spectacularly enough that it, or any number of other things, will have to be cut from the game. That’s just the way it is. You only have 48 hours to slap all this jazz together, and you may want to eat and sleep sometime in there as well. You will try something, it will fail, and you will ditch it. The upside is, you might as well try anything, since who knows, maybe that will be the thing that actually makes it into the game. Once in a while something you think might fail, works so well, and implements so smoothly, that you will wonder why you haven’t been doing it all along. A few years ago I built an entire enemy AI system using Playmaker, a visual state machine editor, and it could not have gone smoother. I think I commented at the time that the AI was about the level of an incredibly stupid roomba. The fact that I had never attempted anything like that before, but it got in the game, and it worked consistently, made me reconsider how I use Playmaker. Going forward I will be doing quite a lot of animating with Skele. Next time I’ll try to make it something more interesting than a hypnotically undulating hand.