“I didn’t know the name at the time, but I always was a bit of a nerdy kid. I was into cartoons, video games, and fantastic worlds. I learned to use a computer and started playing with some graphics software before I learned to ride a bike. When I was 10, I wanted to become a “video game maker”, but didn’t quite know what it was. Grown-ups told me I could choose to learn programming or CG, so I started playing with both a little. Not knowing English at the time made it a bit trickier to pursue programming, so I went the CG route instead.
I started professionally as a compositor in 2008. Little by little I started including programming into my work day, and it now overtook the amount of compositing I do. I have been working for SPINVFX since late 2016, first as a lead compositor, then head of comp, and now head of 2D.”
Note: The following answers have not been edited, to preserve the integrity of Erwan’s responses.
Tell us about the first person or studio who paid you to do VFX. How did the opportunity come about?
In 2008, I obtained my first internship in a VFX studio called Reepost in Paris, where I got to work on some ‘real’ projects for the first time. My internship was as a 3d artist, but that didn’t turn out too well during my first week and so it shifted towards compositing. The team there at the time was just the 4 partners, 3 flame artists and a 3d generalist. I learned so much from these guys. I stayed 3 years there, but eventually wanted to explore the world a little so I moved on.
What piece of knowledge or advice do you wish you had in your early years of working in VFX?
I learn new stuff all the time, and I often feel like if I had known some of them earlier it could have saved me so much time in the past. I guess they call that experience. I wish I had experience when I started out. Now that is not too useful as advice for someone starting out, so I guess my advice would be to get prepared for the ride. There are a lot of up and downs with the VFX life, and that isn’t a great career option for someone who isn’t super passionate about it. You can never really rest on your knowledge, because everything evolves so fast, but if you are loving it, it can get quite rewarding.
As Head of 2D, what tools or processes have you implemented to make your compositing team faster and more efficient?
I share the responsibility of the compositing team with Eric Doiron, Spin’s head of comp. He’s more involved day to day with the team and production, while I’m mostly focusing on the technology side of things, working with pipeline and planning future workflows. One thing we introduced a little while ago was weekly training sessions, and I think the reception of these was quite good. We haven’t really managed to do it every single week, especially in crunch times, but I think it’s a worthy effort.
As far as things that didn’t quite work out, there has been a lot of failed experiments with tools or workflows, but they tend to be fairly minimal. Our leads beta test everything so usually when something is wrong, we catch it early. One failed experiment that is still on my TODO list to fix up was a “smart render” for nuke on our renderfarm, which would automatically retry failed frames in a smart way. My first release of it was a disaster, it started retrying failed frames in an endless loop while putting conditions in place so that it could never succeed. My second iteration of it for some reason killed Nuke’s colorspaces on read nodes. Eventually I’ll get to v3 and hopefully it will be the one.
What is a value or soft skill you wish more compositors would cultivate?
Maybe because of my own inclination toward the technical side of things, I like when artists I work with have some curiosity about how things work. I find that they grow faster once they understand at least the foundations of the math behind the tools.
When you get stuck on a problem, what does your thought process or inner-talk sound like? What’s your process for starting to solve the problem?
I usually start by trying again from the first step, as very often it’s user-error. I don’t know the number of times something wasn’t working because I totally missed a step at the beginning. I like to understand how each step works, so that I can find where things went wrong and why. I google a lot, trying to understand some obscure code error messages that should absolutely not be there, or trying to understand the logic behind a tool or another.
Tell us about your most productive failure — something that didn’t quite go to plan, perhaps in a negative way, that you ultimately learned from.
It wouldn’t really call it a failure, but I would say my stint as a VFX supervisor at Fin Design + Effects in Shanghai for about 9 months could have gone better. I was dreaming of being a VFX supervisor for years, and I did do it once in a while on a project here and there. However doing it full time for a while was a bit of a slap in the face. It was a very small team in Shanghai, and there was suddenly a lot of weight on my shoulders, and I wasn’t quite ready for it. I had always had other supervisors around for support previously and having to deal with everything suddenly was tougher than I imagined. I did my best for a while, and things with projects went fine, but I burned out rather quickly. I’m super grateful to them for giving me this opportunity, and I’m a bit sorry I left so quickly. At that time I just felt I needed a breather. In retrospect, I would probably do it again, but I would do it differently. Now that time has passed I realize I could have made my life easier in a few ways.
Would you be open to elaborating exactly why you don’t think you were ready for the VFX supervisor role at the time (with the benefit of hindsight of course), and what you would do differently if you had the same opportunity presented to you again?
If I was to jump in a time machine and give myself advice at the beginning of my time there, the main ones would probably be to not try to handle everything by myself and be less perfectionist. I tried to do a lot there, just kept on taking a bigger and bigger workload without really letting production know that it was getting too much. I felt some irrational voice that told me that everything was my responsibility. Of course now I realize it wasn’t, but maybe I was too proud to admit I needed more support. In terms of being too perfectionist, I really like to do things “the right way” and was a bit of what we call a pixel f***er. I sometimes spent a lot of time getting the tiny details right, but I was probably the only person ever to zoom in so much on the details and compare before/after on every shot. Of course for some “luxury” clients, it’s all about the details, but I was applying that to everything, and it may have increased pressure on myself and the team for not much gain.
What is the most memorable book, article, podcast episode or TED talk you’ve consumed in the past 12 months (or recent memory), that has made an impact on your life?
It’s not quite in the last 12 months (more like 9 years ago), but reading the art and science of digital compositing by Ron Brinkman really opened my eyes at the time. It was my first foray into the science part, and that is when things started to click. In the last year, I’ve been mind blown by some of the results obtained through Deep Learning. It’s scary and exciting at the same time.
What inefficiencies do you see existing in modern VFX pipelines, and what are you doing to improve them?
On the pipeline dev side, we’re really looking for tedious repetitive tasks that could be automated to save artist time. It can be tiny little convenience things in nuke, like automatically setting paths to save the file and render the images, to developing full on software for specific tasks. Anything that is prone to human error is also a great candidate. It’s easy to waste hours or days because of something dumb, like say a prep task rendered in the wrong colorspace. By removing the human factor, it reduces the amount of error greatly. Of course the tools have to be reliable, because if the tool makes a mistake, it might affect every single thing at the studio.
What advice would you give to an experienced Compositor looking to gain supervisory experience?
I guess I would tell them to make sure they like it. It’s a very different life, there is probably less labor work but more pressure and politics involved. I know quite a few artists who became supervisors, hated it, and went back to being artists. Myself I find supervising a team more suited to me than supervising projects and dealing directly with clients. I like to get to know the people in my team, we spend long stretched of time together and we speak the same “language”. I’m not the best at charming clients. I met some supervisors who can make a whole group of clients smile and laugh in a couple of sentences putting everyone at ease before getting started, where I tend to be a bit too serious and get right into work.
Where can people find out more about you and your work?
I post a few articles on my blog once in a while, it’s a bit sparse because they take time to write, and I’m a bit short on time for that. I’m also reading the nuke forums pretty much every day so you’ll see me active there. I like to take a break from my issues and think about other people’s issues for a minute, it stimulates my problem-solving muscles.