Behind the Scenes: MatterControl
BEHIND THE SCENES: MATTERCONTROL
MatterControl is an original short I made for the company MatterHackers. A while back I made a general video about 3D printing for them, but this video focused on their 3D printing product, MatterControl. The first video showed the vision of 3D printing, and the second video showed how to execute on this vision.
Everything began, as it always does, with the story and the music. I actually write and listen to music concurrently, as they influence each other and drive towards a direction. I spent days listening to relevant, potential music tracks, and it wore me out, as per usual. I ended up listening to 1,487 songs, which was twice as many as I did for the first 3D printing video. Remember, those are unique, relevant songs, so it takes a long time to get to that number.
Music hunting is exciting and inspiring work, but it can also be a bit monotonous.
On more than a few occasions, I found some decent songs, but they weren't quite right for this project. It's a stressful time, trying to find the perfect song and not knowing if it's ever going to present itself. I was pretty anxious during this time, but it finally worked out. Below is an example of a rejected track. It had some elements I liked, but not enough of them. It also didn't have the build I needed at the end, so it ultimately didn't pass my examination.
The story beats were developed in an environment where I have a lot of windows open for days at a time. I'm writing, going over notes from our meetings, listening to music, and doing research on the product. It's the period where I wrap my mind around where the video needs to go.
Day One!
Once my client and I were happy with the story and music, I began work on initial concept frames. At the time I thought I should try to understand the design a bit more before I dove into the storyboards. However, I've since decided that boards and style frames really just need to be developed concurrently.
In the beginning, I have no idea what will work, so I try lots of different things.
In retrospect, some of these styles seem ridiculous, but in the beginning it's all about discovering the direction. I have ideas in my head, but I try to do as many things as possible in case I stumble across something interesting. The client liked the frames that were clean and more minimalist. I then began work on the storyboards, which I sent in the form of an animatic (moving storyboard):
As you can see, storyboards aren't meant to be pretty. Their only function is to get the point across, and begin to test if different movements work. If you compare it to the final version, a number of things were added and removed, but the spirit hopefully remained the same.
Once the animatic was tweaked and approved, I began designing the style in earnest. Building off of my first style frames, I developed what eventually became the look of the video. I wanted it to have callbacks to the first 3D printing video (it's a sister video after all), but I also wanted it to have a unique look of its own.
I spent more time on this treatment.
And so began the final phase of the video: final art and animation. This is my favorite part of any project, because it's when I finally get to make the video for real. This phase lasted about 5-6 weeks, and it was a ton of work.
Characters are not my strong suite, so they're always laborious for me.
Anytime I have a lot of text in a video, I spent a great deal of time making sure every single letter is clean and in its proper place. In this video, I manually cleaned up all the anti-aliasing on every straight edge, and some of the curved edges too. Then I exported every letter individually. Finally, I placed them in AE and set up animation hierarchies. After that all I had to do was animate them.
The reason I do all of this extra work is because I hate how ugly text naturally renders at almost every size. If you look closely at Photoshop's default aliasing, it's garbage. It's no fault of its own, that's just the nature of sizing down a font and then rasterizing it. So, I right these wrongs...manually.
"Text Time" is that portion of a project where I kind of turn off my brain and just get things done. I tend to do this near the end of the day, when my brain can't be super creative anyway.
So, there's a scene that goes by really fast in the video, but it took me many days to create. It's where the tree grows and all the leaves sprout out sequentially. This was all done by hand. I hand-placed 998 leaves. And what with all the null objects and branch assets, there were over 2,000 layers. I had to split up the composition because After Effects couldn't scroll past layer 1900. My memories of this period are hazy, as it was possibly the most tedious scene I've ever made. I used expressions where I could, but there was so much that simply had to be done by me. The leaves had to grow out from the branches in the right way.
Although it was tedious, I kind of like doing this stuff.
Rigging up the woman's walk cycle. Everything is FK, actually.
On the whole, this project was a joy to work on. I pushed myself a lot for this one. Some days were mind-numbing (2000 layers anyone?), but it was all worth it for the learning.
The fine people at MatterHackers are doing great things all the time, too. If you're interested in 3D printing, please go check them out.