And, even more importantly, to start having fun again.
Making your own narrative projects can be hard to pull off, especially when what you're passionate about doing for fun is also what you do for a living. It's really easy to go through long phases where you've kind of lost track of why you got into making films to begin with. It's easy to get locked into a cycle where you lose sight of the fun of storytelling, or at the very least, storytelling for yourself and not just your clients.
Sitting in a Rut
If you're stuck in one of these ruts, my advice is to try and find a new way to look at or approach a story that is completely different from what you usually do to keep the lights on. It just has to feel totally different from your day-to-day work, which for me personally means doc project after doc project and corporate promo after corporate promo, and while I'm happy that I've found a way to make a living in this industry, I decided I needed to find a new way to re-engage with filmmaking.
So, I've gotten really into motion capture animation and Blender/Unreal Engine stuff in the last few years. I became completely obsessed and started writing stories around CG characters and environments, and now I'm happy to report that I'm having fun again.
An Oasis in the Desert
For the first time in a long time, I set out to shoot my own narrative story-based project, but this time I'm using some new tricks and tools to keep the process fun from start to finish. So far, it's been working.
I went to the desert with some friends with the single goal to just sort of be present and have fun while also shooting footage and gathering assets for this story I had written. I wanted to not worry about it. I just wanted to go out there and see what I came back with and put as little pressure on it as possible. While this whole pursuit is absolutely not ground-breaking in any way whatsoever, I found the process interesting because it was an approach that has only been made possible by recent advancements in technology and VFX workflows.
These are exciting times for sure, because this type of project is only made possible now by the recent accessibility of completely free and fully-featured CG and compositing software like Blender and Resolve, free and affordable pipelines and tools for motion capture animation and character rigging, and super cheap methods for scanning 3D assets and textures. Here are some reasons why I'm excited about trying out these new approaches.
Credit: Logan Baker
VFX That Were Once Insanely Expensive Can Now Be Done for Free
Peter France is a stalwart of the Blender community. He's always been a strong advocate for the software, and for good reason. For as long as I've been following his work, he's been able to achieve amazing results in shorter amounts of time than just about anybody else. The most recent example is his short film Scooty.
In the above video, Peter goes over all the exact processes and tools I mentioned before. All of these things are available for free (Blender, Resolve, Polycam, etc). Peter made an 18-minute short film by himself that features a photo-real and extremely well-animated CG character in almost every single shot of the film. This is such a perfect example of the sort of thing I'm talking about. Setting out to make a film like this even four or five years ago would have been either very expensive or simply impossible pursuit.
Motion Capture Is More Accessible Than Ever Before
Motion capture used to be the work of an entire team of engineers. You'd have to build elaborate truss systems with light-sensing cameras that would then be patched into multiple computers for real-time gathering of the motion capture data. The whole thing was obviously very expensive and was pretty much cost-prohibitive for anyone that wasn't a major studio or game developer.
Now, if you want, you can spend $3,000 to $4,000 on a Rokoko Smartsuit Pro setup and get started. That may sound like a lot of money, but considering the next option up is a $20,000 to 30,000 Xsens suit—it's really a great deal.
The Smartsuit Pro is great, and the support team and company behind it is great. There's even a really active community and Discord server that you can go to for quick help when you run into a snag. Trust me, you will. I've never once set out to do motion capture and had everything work exactly right on the first try.
But when you compare the time spent on that versus hand animating, it's worth the trouble.
If you don't want to fork it over for a mocap suit, some emerging free options are super interesting. Plask, for instance, allows you to drag and drop a video file with someone moving around in it, and it will use some nifty AI to extract the motion from the person in your video and apply it to a rigged character.
I've played around with it a decent amount and I gotta say, it's really fun and fast, and represents an emerging part of the mocap world that will be super dangerous in a mere couple of years. I'd wager that within the next two to three years we'll be able to just pull out our phone and instantly record mocap data and apply it directly to a character in our CG software of choice via the cloud.
You Can Make Photo-real CG Assets with Your Phone
While I was in the desert, I spent a lot of the time gathering assets that I might need to do the VFX in the film. One example of that was using Polycam to create 3D models of rocks and other desert features that I knew I'd end up using. Scans like this one.
These scans have become such a big part of my VFX workflow. In certain moments, using your phone, you can have an extremely highly detailed model of just about anything that isn't shiny. It almost feels like cheating half the time.
With Polycam, you can also take scans of your locations while on-set and use those scans in conjunction with your shot footage to get completely accurate lighting in your renders because shadows are created where they should be, and light will be bouncing off of the correct colors. Peter mentioned using this process in his Scooty BTS video above. The potential uses for this are endless.
Everything Is Faster All The Time
In just the short time since I've been a Blender user, advancements to the render engines as well as faster GPUs have reduced render times by around 300% or more. Stuff that used to take 1 minute per frame now takes 10 seconds.
Pair that with the upgrades to the de-noising tools and you've got lightning-speed render times. These types of changes aren't specific to just Blender, as almost all major render engines have real-time options now, and almost everything is making full use of GPU acceleration. Render engines are all just faster than they've ever been. Don't even get me started on Unreal Engine 5's mind-bending nanite technology.
I think I talk a lot about how fast times are changing, but it's just so true. Within the next decade, everything could change or look entirely different. I'm just really trying to focus on making sure that I'm using all of these new tools to continue learning and have fun making movies again.
There's always going to be something new to learn, and then take what we learn onto set and use it to challenge and change our filmmaking process every time.
