Gareth Fowler talked about his career in graphics, discussed his approach to mechanical animations and shared his favorite software for its production.
Introduction and Career
Hello! My name is Gareth Fowler, I'm a Graphic Designer, Illustrator, 3D Artist, and general all-around visual guy. Currently, I work for a small boutique branding agency in Toronto named Haft2, where I'm the Senior Production Designer. Mostly, I work as a traditional Graphic Designer but, occasionally, I get called on for my Illustration and 3D skills.
The mechanical work started maybe 4 or 5 years ago. Long story short I was just bored out of my mind with graphic design as I had been taught and the limitation of the software. Photoshop, Illustrator, and InDesign might have been my bread and butter, hell, they still are, and I love them dearly, but, dear lord, can that get boring after a while. I'd played around with 3DSMax, Maya, Blender, Sketchup, and probably a few other pieces of software here and there, but eventually I got my hands on a janky old copy of Cinema4D, and I was just totally blown away with how easy and intuitive it was for me. I started with really simple and definitely more abstract designs and animations to start, real retro wave stuff, and that was all well and good, but I wanted to do something more, something really out of the left field. I know there's this predilection among designers, regardless of their field of expertise, to stay in their lane and pull inspiration from others in their line off work, and there's nothing wrong with that, but you do run the risk of getting caught up in a visual echo chamber. This stuff is cyclical. I've always loved cars, motorcycles, planes, anything loud and mechanical really, and I've always found myself going to that place for inspiration in my own work. When I started teaching myself rigging, I realized that, yeah, sure, I can rig a walk cycle, and that's awesome, but that also means I can rig a simple 4-stroke engine, and that was way more entertaining to me.
About the Idea of Projects
As I said, I'm into anything loud and mechanical, and I don't view firearms any differently than I view, say, an engine. It's an object, it has no political agenda or ideology, it just is. And as it's a complicated mechanical device (at times), it is incredibly interesting to me. The firearms series wasn't exactly a single, unified concept in my mind so much as it was an extension of what I was already doing. I had done a few other mechanical animations, mostly automotive based, and firearms were, in my mind, a logical extension of that, insofar as they were basically just objects powered by explosions. That's pretty metal. It was also a good excuse for me to make slightly more complicated animations, with even more moving parts and considerations. I liked the challenge.
Gathering the References
The first step is usually seeing if I can dig up some patent drawings or a YouTube video or something, anything that will help me understand just what is going on in the mechanical heart of whatever I'm thinking of. From there, I go and make what is kind of like a skeleton of splines and nulls, no modeling at this point, and I animate those. It's a way for me to quickly prototype an idea and get all the mechanics down. I've attached a short video of what this looks like for a simple piston engine. I don't much care for accurate, 100%, one-to-one reactions of action or mechanical principles, and that's not my goal, anyways. My goal with that series, the firearms anyways, was to show just what was going on mechanically without resorting to transparent objects or floating parts or whatever. This, of course, necessitated taking some creative liberties with the original designs, some of which would likely be ill-advised if they were ever reproduced in real life. I liken it to a stripped-down, open-wheel race car, with all that exposed suspensions, the bodywork being 100% functional and aerodynamic, and it just screams deliberate intent. Now, is that practical for day-to-day use? Hell no, but it looks cool!
Oh god, I am probably the last person who should provide modeling advice. I'm totally self-taught and probably have enough bad habits that I've pretty much forever talked myself out of doing it professionally full-time. nGons for days, but I'm getting better at that. Having said that, I consider myself nothing if not disciplined, at least in my own weird way. I'm big on round numbers, consistent dimensions, and keeping a grid in mind. I'm sure that's my background as a Graphic Designer exercising itself in a new medium but hear me out: being consistent saves times, offloads a lot of creative thinking, and it allows you to reuse parts and elements. It keeps things on a similar scale and will let you know real quick if you've screwed up. Also, it results in a unified visual language that can be continuously applied over a long period of time and across a variety of designs. I might change my lighting rigs or textures here and there, but if the models maintain some consistency, if there are common visual elements used across all of them, then they start to look less like one-off designs and more part of a larger body of work. Or so I tell myself. Oh, and I do all my modeling within C4D currently, but I've been playing around with some parametric modeling in Fusion360 and I'm liking what I'm seeing so far.
Working on Animation
I mentioned earlier about the skeleton of splines and nulls (I've attached a video illustrating what I mean by that), and that's really where it all starts. For the most part, I never place any real geometry until I have the basics all figured out, keyframed, Xpresso-ed, whatever. One self-imposed rule I try to stick to is to have as few keyframes as possible. For example, in the wireframe attached, the only keyframes are at the beginning and end, and then only on the crankshaft. Everything else is driven off of that through constraints or Xpresso. Once I have the skeleton done, I start adding geometry, usually one piece at a time, again with an eye for consistency and repeatability.
I like to keep it simple. All the materials are usually 100% Reflectance channel, maybe some Luminance or Bump work but that's it. The lighting is usually the simplest HDRI I can find, I like simple light tents, and maybe an additional area light here and there for highlights. Currently, I'm doing all my rendering entirely with C4D as I haven't had the time to learn RedShift or whatever another flavor of the week GPU rendering engine is making the rounds. The neater and tighter I can keep my workflow, the better I feel about the final product.