Thanks!
I did start with 3 advantages: Conceptually I was already familiar with key-framing and virtual cameras because I’m an editor. I also help develop live-action fight scenes. And I’m versed in cinematography.
But TBH, the secret behind my initial shorts coming out pretty is…I didn’t plan them out.
Usually, planning is so important. Especially for collaborating.
But since I’m learning, I sorted out the action of this short film as I went along, shot by shot.
I started with good-looking assets from the Marketplace (which happened to be free). The Paragon characters came with a bunch of animations. I found an attractive angle of the city and had the idea for a character on a beam in the foreground. That was the only idea I had in the beginning.
So I moved a beam into place. And put Phase on it. Then I cycled through her animations and settled on her running along the beam. It became obvious she would need to stop or jump. So I figured out the jumping concept and duplicated a bunch of neon signs to be the scenery for her plummet.
There were a few landing animations available and the one I liked most - where she kind of stumbles to a stop - I realized there was a moment that could be interpreted as her encountering danger, if I cut off the animation early.
I knew I wanted conflict with an antagonist, so it was a good moment to put in a baddie. I looked at a bunch of Paragon characters and settled on Crunch, but I changed the materials on his body to my liking. And I scaled him up. For the initial encounter, I once again cycled through animations, but this time I needed moves for two characters that played together. I knew Phase had to be more agile than the hulking robot, so I found animations that kind of looked like dodges. They weren’t actual dodges, but by messing with the animation timing and her location, I could sell them as dodges. When she jumped in the air…at first I thought ’ this is where the balance of power will flip.’
But then I had the idea for her to be punched through the air into a neon sign. All the other details - like the first sign she chips off, and the signs that fall later - that stuff was figured out in the moment. The first sign she hit was just in the way of her path to the main sign, so I decided ’ let me see if I can sell this chip (bouncing off the edge of something)’ instead of moving the first sign out of the way.
And that’s basically how everything went. I looked at the map and figured out cool backdrops for the action, then figured out the animation, then figured out cool camera movements. Based on camera angle I did the lighting.
By using what I considered to be the best available assets and animations at every turn, I was able to create a pretty slick project, because I was leaning on the valuable creative work done by experts. I was just combining it all together in a unique way, incorporating my background as a fight scene guy. Then applying my strengths in camera work, and doing lighting (which I’m not an expert in but I understand the basic concepts of lighting for film).
However, as this short took shape, I realized I’d need some other animations that weren’t included with the characters in order to bring the story home. So I had to learn skeleton re-targeting so I could use third party animations, and I also learned to do some custom animation by hand (Phase writhing on the ground after falling).
The re-targeting was actually the hardest thing for me to learn, harder than doing original animation. Because re-targeting simply has no equivalent to live-action filmmaking. Everything else had an equivalent to directing or editing or DP’ing. But re-targeting was new, alien, and confusing to me.
Other things that have been difficult for me to learn include socketing a hat or weapon on a character. It’s just not relatable to live-action filmmaking. Also, obviously if hair physics was an option, I didn’t know how to turn them on. I’m still learning so much.
You asked about storyboards. Initially I got into UE4 just to use it for rough previz storyboards. This short films wasn’t planned out - instead, I approached it more like previz, figuring it out shot by shot. Maybe I can call this approach “action stream of consciousness.” Eric Jacobus - a great stunt guy, mocap artist and filmmaker - suggested I learn UE4. All I have to use is a 2015 iMac, so I figured I could only do minimal things with the engine like storyboarding. Although I desperately want a big powerful PC with RTX for Unreal, I’ve still been pretty impressed with the amount I’ve been able to do on this old iMac.
