Blendering

Posted by Rampant Coyote on December 13, 2010

I’ve been using an old-and-creaky version of Blender (around 2.43) for a long time, now. I had an exporter I knew worked for my engine (which is no longer being maintained), and I was pretty used to how things worked. I finally upgraded to version 2.49 – which is getting a little bit old-and-creaky on its own, but AFAICT that is the last version my exporter supported.

It took maybe a half-hour for me to get it set up with the last known update of the exporter, and to test things out to verify they worked. Maybe another hour to learn new ways Blender 2.49 does things over the older version, and to explore the differences in how the exporter now expects data to be formatted. There may be more to know, but it for now I’m productive.

And then I spent about another half-hour or so re-learning stuff I thought I already knew. As it turns out, there are far superior ways of doing things, like UV mapping and rigging, than I’d been using. Techniques that apparently existed even in my creaky-old version. Techniques that probably would have cut twenty or more hours out of my schedule had I taken the time to learn them two years ago.  That may not sound like much, but that’s a very good week of part-time game development for me.

So I guess the lesson is that while there’s no training as good as just hands-on doing it when it comes to game development, you do need to take some time for additional learning. Especially with tools as complex as a programming language or a 3D modeling package.

What keeps occurring to me is that I need to be better on the art side of things. While I may be able to get by to a degree with stock off-the-shelf content and a few horrendously underpaid contractors building up their portfolio, I still need to be able to do it myself when push comes to shove. Why?  Well, let’s see… I’ve had one artist disappear on me, and another one that’s had massive Internet and real-life issues. Content purchased off-the-shelf is rarely exactly what you need, and will often need customization to fit your game. And even if it’s a team member who is actively producing quality stuff, oftentimes there are quick, minor changes that may need to be made at midnight – like moving an origin or mount node around, or fixing a minor texture bug – and it is easier to just make those edits on the spot. It helps to be able to speak to an artist in their own language, as much as possible, so being fluent in the issues they face getting a model to look and work right is handy. There are times where you just need something small done quickly and can do it yourself, or when the rest of the team is overtasked and you can take up a little bit of the slack.  And finally, as I’ve said before, an indie should be able to carry the project on his or her back if necessary for whatever reason.

I still need a lot of practice.

I struggle somewhat giving myself time to practice, as I realize that for the immediate future, it’s not the most efficient use of my time, and my results are sub-par. But at least I can see that I’ve improved a lot. So something is working. I clearly suck less. I’m always about sucking less…

I’m not applying the same logic to music, though. At least not yet. One thing at a time.

So that’s my excuse for spending all my dev time this weekend working on something that looks like a cross between a manta ray, an alien face-hugger, and the Starship Enterprise. It was all about learning, see. And practice. And maybe, with about four more animations and a few more changes to the model, replacement of stand-in content with final art.

Filed Under: Art - Comments: 4 Comments to Read



top