Episode 85 – Interview with Philippe Leprince, Senior Field Engineer at Pixar Animation Studios

 

Click here to listen on iTunes!

Get on the VIP insiders list!

Check out www.VFXRates.com

 

In this Episode, Allan McKay interviews Philippe Leprince, Senior Field Engineer at Pixar Animation Studios. He currently works as an Engineer on the RenderMan Software. Philippe has a Lighting and CG background and has worked at Framestore and Double Negative, and many other big studios. Among his many credits are feature films like Harry Potter and the Prisoner of Azkaban (Framestore); The Dark Knight and Inception (Double Negative).

Allan and Philippe discuss the benefits of being a well-rounded artist, the importance of small details, tips for having a successful job interview and much more!

 

Philippe Leprince on IMDb: http://www.imdb.com/name/nm1473896/?ref_=nv_sr_1

Philippe Leprince on Twitter: @rendermaniac

Philippe Leprince’s Blog: http://leprince.co.uk

 

EPISODE 85 — Interview with Pixar’s Senior Field Engineer Philippe Leprince

 

Hi, everyone!

This is Allan McKay. Welcome to Episode 85. I’m speaking with Philippe Leprince from Pixar. Let’s dive in!

 

FIRST THINGS FIRST:

Lot of things are going on at the moment.

I. [-1:03:02] The free training that I’ve mentioned (allanmckay.com/fireball) is still going for a week or two. This a 15-hour, high end VFX training course — for free! We cover a lot of stuff:

– digital pyrotechnics;

– disintegration effects;

– rendering;

– compositing;

– lighting.

If you want to check it out, go to allanmckay.com/fireball.

[-1:01:48] I’m doing a lot of live training right now. You can check those out of Facebook Live: www.facebook.com/allanfmckay. We’re doing:

– feature case studies;

– live courses;

– a lot of stuff behind the scenes.

 

[-[1:01:09] In this Episode, I’m speaking with Philippe Leprince from Pixar. He’s a Senior Field Engineer in the UK. He has a Lighting and CG background working for Framestore and a lot of other studios.

Philippe: Currently, I’m a RenderMan Engineer at Pixar Animation Studios. I’m working on the Maya plugin at RenderMan.

[-[1:00:51] Allan: That’s awesome! I’d love to hear about how you got started. I have yet to hear, “It was easy.” How did you get started? Did you always want to work in film?

Philippe: Not really. When I was young, there was no CG, really. My first motivation was from being into science fiction, as a child. One day, my grandmother took me and my brother to see Star Wars.

Allan: It’s always Star Wars and always Jurassic Park

Philippe: Yes. I was eleven, or something. It was completely different: It wasn’t a cheesy science fiction movie. It was believable and realistic! That was an important moment for me in retrospect, but it’s not my favorite movie. I was more influenced by Ridley Scott: Bladerunner and the first Alien.

I wasn’t very good at school. I was lazy and always taking the short route to get to the next level. I was taking one class to get to another. When the university time came, I didn’t know what to do. At the time, I was into music. (I had a band with my brother.) I was keeping an eye on computer graphics, which were still pretty new. There was a computer graphics conference in France called Imagine. It was the equivalent of SIGGRAPH, but it focused more on images. It was more about the image and the artistic aspect.

There was no school or university offering any [computer graphics] courses at the time, at least in France. After studying to be a technical translator for a year, I dropped out and had to do some military time and then started working as a photo touch-up artist. That’s when I started using 3D for real. I had played with an Amiga, one of those things on a floppy with 4,000 colors. I played quite a bit with it, but it couldn’t do much, really.

When I was working for the first company, I had an opportunity to make my first 3D images. One day, I stumbled on RenderMan during a conference in France.

[-[55:55] Allan: What year was that, by the way?

Philippe: I’m not sure.

Allan: Was it around Toy Story in 1995?

Philippe: It was prior to that. They decided to release a MacIntosh version of RenderMan, version 3.4. That was probably in 1999. I was playing with that and I was completely hooked. At the same time I was learning compositing with the first version of AfterEffects. All those things came together. I started doing commercials with little resources, but it went well. After a couple of years, I moved on to a bigger CG company in France called Deus.

[-[54:06] Allan: What was it like going from one company where you’re doing 3D to another company where everyone embraced it?

Philippe: Yeah, at the first company, doing 3D was seen as something exotic; but that wasn’t the main purpose. It was nice to have a flying logo or something; but no one thought it was important. Deus was a much more professional outfit, with graphics. It was really fantastic to move from using a Mac to Silicon Graphics; getting access to PowerAnimator, which was the best solution around [at the time]. I also realized that Mac was able to do a lot of things: Photoshop, a lot of things you couldn’t do with Silicon Graphics (not without shelling out a lot of money). We had 8 work stations. It was about making it look good but making sure it was light weight [when it came time to render]. We worked mostly on commercials. After that, there was a bit of restructuring [in the industry] in France at the time, and Deus didn’t make it. I started working as a freelancer.

[-[51:04] Allan: What was that like to transition to freelance? Was it pretty scary?

Philippe: It was, a bit. I ended up visiting a bunch of companies. I got some work pretty quickly. There weren’t many people using PowerAnimator, or even the first version of Houdini at the time. Those were the two things I could use. Overall, I was able to find some work pretty quickly. I started to get a good reputation as someone who wouldn’t be stopped by anything. Because I always had this approach of doing things in a light weight manner. Working for a small company with MacIntoshes, or having three machines to render things [taught me that]. You have to be careful about everything, do some parts in 2D. That was really about being as savvy as possible; not wanting to be too correct, but making sure that everything works at the end.

[-[49:11] Allan: Which, to be honest, is a good way to start out. I always think that these days we get really spoilt with all that render power, and everyone doing little bits and pieces. Whereas, if you go from being a Generalist at a smaller studio, you have a lot of responsibility, a lot of the times it is about the cheats and hacks. In a way, it allows you to think on a larger scale later and much more efficiently. Same thing if you come in as a Modeler, you will be thinking about things down the line. 

Philippe: Later on, when I was working on bigger projects, it [was] important to divide and conquer. It’s good to have a good analytic outlook on what you’re trying to achieve and split it into a bunch of manageable tasks so that you can deliver on time, with not too many unknowns on the way. Also because I had to rely mostly on myself for the small things — I had to learn lighting, shading, texturing.

When I started working at Deus, I started looking into basic programming skills. My first computer [was] an Apple 2 that I was sharing with a friend. We started learning BASIC. I realized even if you buy really expensive software on Silicon Graphics, it didn’t have all the stuff you wanted. There was SDK. You could program your own stuff. RenderMan was a big factor for me. You could write shaders really easily. I was able to experiment a lot. Later on, it was another thing in my favor. 

It’s really important to be very flexible and know more than one thing. I think a lot of people try to specialize a bit too quickly. I interviewed a lot of people when I was at Double Negative. Often time, they were already way too specialized. They wanted to do Modeling or Lighting, and nothing else. You can’t do just one thing in your life. You need to have a wider experience because it’s enriching your outlook.

[-[44:58] Allan: I think that’s so critical. If you don’t understand where your stuff is coming from or where it’s going, it’s going to be really hard to make things right for the other person. We’re all part of a pipeline. 

– Start as a Generalist;

– Understand where things go through Production;

Then start to niche down, after a couple of years. 

Otherwise you’re shooting blind. These are the basic things we need to know.

Philippe: I think it’s also part of the fun. All the stuff you learn all the time — it’s really empowering. Trying to get to the technique, that makes your better at everything.

Allan: And you might discover something you’re really passionate about you wouldn’t have discovered otherwise.

Philippe: Exactly!

[-[43:22] Allan: Just touching base on that point of interviewing artists. I always find that interesting. What are the common things / red flags that would come up that would make or break hiring them?

Philippe: When I was interviewing people, it was mostly around shading and lighting.

I. But I think in general, it’s important to find someone who is nice because you’re going to work with that person. There is that technical side of things, but the personality comes across too. It happened a few times: The person is a bit cold and you’re not sure if you want to work with that person.

II. You need to connect with the person who is interviewing you. As an interviewer, I’m here to find somebody. If you’re the right person, I would rather take you than have to interview 20 more candidates. I think it’s important to let your personality come through.

III. When it comes to the Demo Reel, it’s important to care about what you’re showing. I’ve had people who’ve worked on larger projects. They weren’t entirely honest about what they worked on. We realized that. Be honest and if you’re passionate about it, it’s totally fine. All the companies are looking for people who can do things well, but we realize we can’t always find people [of senior backgrounds]. We need to have people who are new to the game and willing to learn, get better at things by working alongside other people. It’s really important to be honest and reflect at your level without feeling diminished. You’re not supposed to be the best about everything.

[-[39:58] Allan: If you’re embarrassed and lie, it’s such a small industry it’s not worth pulling wool over people’s eyes. You don’t want to steal people’s work and gain that reputation [either]. It’s not worth it at the end.

Philippe: People know each other after a few years. There are still a few people in the industry who are hopping from company to company, but they manage to sell themselves.

[-[37:24] Allan: Do you think it’s important for all artist to have understanding of basic scripting? Do you think that puts them into a different category?

Philippe: Yes, I do. The more you know, the better you are. I always find it disparaging when people say that they are artists, [not technicians]. If you look at Leonardo da Vinci or Michelangelo, they could do so many things because they were curious. It’s just a general attitude toward life. If you limit yourself or get scared of the idea of getting down and dirty with technical stuff — take baby steps.

[-[35:41] Allan: Most of the time when you’re coding, it’s one line; then you start expanding and pulling other bits. It’s a jigsaw puzzles. It’s something that you keep building. I think you’re right there is a massive intimidation when it comes to scripting. The best artists I know, they all code. When they hit that wall, they’re able to build over it. It’s so critical! When it comes to production, you become the go-to person in the room.

Philippe: If you have special skills, people will know that. There is always a point when you need to do something that has not been written by Autodesk, or whatever. That moment is between you and the computer and there is nothing that can come in between that. It will become second nature. If you can do it in a company with other like-minded people, you’re going to come up with more ideas. That’s what I loved when I worked at Double Negative: There was such a strong group of senior [artists], when we were in a bind, we would have a brainstorming session and go find a solution. They were all guys that could script; write plug-ins. You don’t need [to be at] an amazing level to have a good idea.

[-[32:22] Allan: You mentioned when you got started, was it something you felt compelled to learn. I realize that back then you were forced to do that.

Philippe: It’s kind of a funny story. This friend I was learning with — he is now a co-director of Despicable Me — we wanted to play video games on a computer with BASIC. So we wrote some code where were could animate character. Took us a few weeks, and that was our first motivation.

[-[30:52] Allan: I do think that when you’re learning, it’s important to have a goal in mind. Having a goal is critical because that way you have a direction. If you’re just saying you want to learn coding but have no way of translating it, you will have no direction. 

Philippe: My friend Danny, when he sees a problem, he smiles. That’s the fun of the job: the problems! If you don’t want any problems, don’t go into VFX.

[-[29:24] Allan: I agree. This industry and this job is purely problem solving. When you worked at Framestore, that was a big leap from working in France. What was that switch like?

Philippe: They were just starting to be called Framestore CFC. On a nice Thursday, I took a train to London and visited MPC, Framestore, Double Negative and Cinecite. In one day, I managed to do 4 interviews. I had a proposal from Framestore a week and a half after the interview. It was probably the most amazing interview I’ve ever had. I went in thinking I’m just a TD who can do a bunch of things. It was a big company in comparison to what I was used to.

There were six supervisors [in the room]. They had all seen my reel and they all wanted me to work on each of their projects. That was really fun! I ended up working on Harry Potter and the Prisoner of Azkaban. I’d done motion control work on commercials. I was made in charge of writing the motion control pipeline. That took 11 months. I spent 9 weeks on set supervising. That was very successful. I stayed at Framestore for 3.5 years.

[-[25:34] Allan: I love that. Let’s say you come from a small industry, you just don’t know how you’re going to [fit in] with the big boys. It’s always reassuring to see the reaction of others at your skills. 

Philippe: When you start working at those big companies, you realize there is a range of skills. I had a few friends working at Pixar, ILM and they were telling me to come onboard. That was the best thing that could’ve happened to me.

[-[23:53] Allan: In terms of the projects you’ve worked on, what were some of your favorite?

Philippe: My two favorite projects I’ve done at Double Negative. They were: The Dark Knight (just awesome!) and Inception. Two-Face at the hospital was done by Framestore. Double Negative did all of Gotham. The Hong-Kong sequence was done by Framestore; the rest was done by Double Negative. Every time you see things happening in Gotham at night, that’s us. A lot of Batman’s stunts. A lot of replacement of Batman himself. It was a really fun job! It was the first time we did anything in IMAX.

[-[22:41] Allan: To just add that extra bit of pain!

Philippe: Yeah. You only have to render at 6.5K.

Allan: Was it stereo?

Philippe: It was not stereo at the time. We were using Shake software to composite. Shake wasn’t able to fit everything into memory. So we took the code from Shake that we had acquired from Apple, to bring to 64 bit, to render those images and to comp everything. Once we had these 64-bit Shake, we were finally able to get the images but we couldn’t see them because there was no way to project them. So during the dailies, we would see a resized image and we would look at the image in sections. You had to have six sections for every sequence. It was painful and slow. Sometimes we would go to the IMAX and we would watch it over there. That was a great memory!

[-[21:00] Allan: I’m sure! Although at the time it must’ve felt ridiculous to see it in tiles. It’s pretty cool to work on projects that have that challenge. I used to love Shake! What about Inception? What type of work did you do on that?

Philippe: We did everything on Inception. At the beginning, I was a Rendering Sup. I was in charge of deciding how we would pipeline the show. After a few years, my specialty became rendering stupid amounts of geometry and making sure it would render on time. On Inception, we again had cities, and a beach environment, clean-ups and replacements at the hotel. In the beginning, I was involved in the Paris folding sequence. That was sequence was supervised by Daniel [Baldwin] with the help of Bruno Baron. A fantastic team as well! There was the beach sequence and the dead city sequence. Those were the two main sequences I’ve worked on.

[-[18:49] Allan: What were some of the challenges on that project?

Philippe: There were just complex in the terms of the amount of stuff we had to render. Every time, you have to render a full city, it’s a bit of a challenge. There are always those camera moves across 20 blocks and those have to look good. There is complexity in terms of storage. My work consisted of writing the shading tools and making sure to track the performance. There was a lot of back and forth with the artists. It’s really interesting! It’s not as artistic as the stuff I did in the beginning of my career, but I did shots as well. I was able to eat my own dog food which is really important when making new tools.

[-[17:18] Allan: Yeah. You’re right about that. How does it feel to go from working on shots to be more of a support for other artists? You’re giving them the keys to do everything they want to do.

Philippe: At the beginning, I was interested in making pictures. After a while, the general frustration about the state of the tools we could buy became more important than the artistic aspect of it. I became completely obsessed with improving the tools. Artists need good tools to do good work. Sometime, it’s the small details that matter. They will make a difference. So it’s important to have that attention to detail. In the same way you must have an attention to detail when you’re lighting, doing shading, texturing. It’s how things become 10-fold better. When the tool has evolved to do what it’s supposed to, now you can go the extra mile. The most difficult part is that last 10 percent.

[-[14:58] Allan: You’re right. I always feel that all the tools out there are really general. It’s not laser focused. That’s why learning to script can help you design or redesign for every production. 

Philippe: You’ve worked for different companies, and you know that every company is using pretty much the same software, but they use it differently. That allows them to do something better than the version everybody buys.

I think sometimes people are too impatient. They want the solution immediately. But I think you learn more when you fail. If something doesn’t work, either you give up and lose. If you don’t give up, you finally discover why it’s not working and you’ll be able to use it in the future. There are too many people focused on the end result. It’s not the destination, it’s the trip to get there. As an artist your main goal should be to get better at what you do, and that takes different forms.

[-[10:59] Allan: Absolutely! I so firmly believe that in my core. A lot of people just look at what buttons to press as opposed to why it’s happening. If something is not working, go back to basics and make it work. It’s basic formula. Having that knowledge allows you to not hit walls. 

Philippe: I think it’s possible to explain complex things in a simple way. If you have an understanding of a complex problem, it’s easy to explain it to someone else. It happened to me many times. I could tell which guy was good and which guy wasn’t applying himself.

[-[08:04] Allan: Now, moving on to Pixar. You’ve been there a couple years now, is that right?

Philippe: Three years now.

Allan: And how is your role different there?

Philippe: Basically, I’m in the RenderMan team. We’re writing RenderMan. Three and a half years ago, we delivered the first retracer in the line-up. That was RenderMan 19. I was hired to help to make it production ready. That’s what I’ve done. I’ve written a large section of the shading library. It’s about making tools that make sense, but it’s difficult. We want to have a render that’s non-bias. We’ve had a lot of people say it’s complicated. They have a point. It’ll just take some time and innovation on our end. We’ll present new stuff at SIGGRAPH this year. With version 22, there are huge improvements for us. We’re getting to that point where everything is more clever. It means you get good results pretty quickly.

One of the things we didn’t like about the caching system was the fact that you had hundreds of sliders and knobs. Out of the box, it wouldn’t give you good results. If you do a single frame, it looks fantastic. That’s not the way a render should work. It should be like an iPhone camera that gives you great pictures as much as possible, depending on the lighting condition. We want to get to the point where you set up your lights and press the button — and you get a good looking image.

[-[04:03] Allan: Just at this moment, I remembered when I uploaded RenderMan. It was in ’98. I brought in an object expecting it to look amazing. It always comes down to the artist. Do you still have a lot of interaction with studios, in terms of support?

Philippe: Yeah, I spend a week every month at ILM. We visit other studios when necessary. I still go up to people’s desks and grab their mouse and show them how things work. Our team is pretty small. We’re only 20 people. We try to do as much as we can. Sometimes, I have to explain thing because we don’t have the documentation for it.

[-[02:14] Allan: If people want to find out more about you, where should they go?

Philippe: Well, there is Twitter: @rendermaniac. I mostly post stuff about RenderMan. I also have a blog at http://leprince.co.uk. I sometimes post images there of tests and things.

Allan: Thanks for taking the time to do this! This has been so awesome!

Philippe: My pleasure!

Hope you’ve enjoyed this Episode. Again, I want to thank Philippe for doing this.

I will be back next week with another Episode. It will be the last week you’ll be able to get the training at allanmckay.com/fireball. Get it while you can!

Rock on!

 

If you enjoyed this episode, please share it using the social media buttons you see at the bottom of the post.

Also, please leave an honest review for the Allan McKay Podcast on iTunes! Ratings and reviews are extremely helpful and greatly appreciated! They do matter in the rankings of the show, and I read each and every one of them.

And finally, don’t forget to subscribe to the show on iTunes to get automatic updates.

Let's Connect

View my profile on

INSTAGRAM

wp_footer()