If you have read any of these posts before, you know that I have a serious love/hate relationship with Blender. For all it’s advances and additions over the years Blender still remains one of the more inscrutable tools I have ever used. That list includes the gloriously impenetrable ZBrush and some proprietary sign manufacturing software designed by moderately intelligent squirrels. I have used Blender to rig characters before, but had you interrogated me afterward, I would have conceded that I may have been aided by wizards. Mind you, I’m not new to character rigging. I have written code for custom rigs in Maya and Max, so even at a base level, character rigging is a concept I’m familiar with.
This week I participated in a series of cage match battles with the Blender rigging system. I’ll be honest, I wasn’t sure which of us would exit the cage. I eyeballed the Maya LT license page, aka “the tapout” more than once. Going the Maya route would be more comfortable, possibly faster, but it would also cost more money than none, which is what I currently pay for Blender. At every roadblock I had to ask myself if learning the obtuse way that one or two programmers thought these tools should work was worth the effort. It’s like a professional level adventure game puzzle. I expect that Blenders Inverse Kinematics solvers will incorporate a rubber chicken and some chewed gum at some point.
This weeklong fight allowed me to pinpoint my problems with Blender. It turns out they are very much the same as the problems that I have with Apple products. The same “simplicity” that most people enjoy typically means that anything under the hood is obscured from the user. When I search around for an explanation of different tools in Blender, I find a lot of people saying feature X works this way “just because” Unfortunately for me, I have the sort of brain that is not, and will never be, happy with “just because”. It’s probably why both Maya and Unity clicked for me. Both of those programs bombard the user with information about what is going on in front of them and just say “here, deal with it”. They try to put the velocity of the information hose under the users control, but the feedback loop of your actions resulting in visible, recorded changes, is tight enough to tell brains like mine “here’s why”. I can learn systems like Blender too, but it might take me a week.
That said, I still remain positive about the future of Blender. There seems to be a serious interest in making most features nodal, which would would improve usability and visibility of buried features. An actual viewer for the dependency graph would also be incredibly useful. These features would go a long way in breaking the hobbyist stigma that still clings to Blender, despite it having all the features of most of the pro 3D programs. Right now Blender is in a race though, whether they know it or not. If Maya comes out with a much cheaper or even free option, before they can get their act together, the race will be lost.