Peter Demarest - Compositing Supervisor for Boba Fett and The Mandalorian - now at Pixar
This week I chat with Peter Demarest, compositing supervisor on Boba Fett and The Mandalorian, as well as working on some of the most iconic shots and movies for the past decade, such as The Avengers, Avatar, Jurassic World, and the Star Wars films. I saw Peter's reel a few weeks back and was just blown away, so wanted to reach out and chat to him, and here's that chat!
Take a look at Peter's compositing showreel - there's a link below. I loved this chat, we talked about Peter's career and some specific films and turning points in compositing, and also how the role has changed and is changing with virtual production on projects like Boba Fett and The Mandalorian. Peter's recently started at Pixar, so we also touched on how the role of compositing is changing in animated features.
Regardless of whether you think you're interested in compositing or not, I encourage you to dig into this episode, because I always find that I learn the most from episodes I least expect to, not just here, but in other podcasts as well.
Peter's Compositing reel
Peter's IMDB
Please remember to like, rate, and comment on your favorite podcasting platform and share the episode on social media.
If you have any comments or suggestions please get in touch.
Host & Producer: Michael Wakelam
Executive Producer: Eric M. Miller
Music by: Rich Dickerson
Audio Engineering: Mike Rocha
The Creators Society is a professional society for all disciplines of the animation industry. Our mission is to bring the animation community together to build strong relationships, provide education, and form a better understanding of the different roles we all play in creating animated stories. We celebrate and promote the love of animation, and all the talented Creators who breathe life and imagination into their work.
Episode Transcript
Peter Demarest: 0:14
Star Trek gosh that was an incredible film and really influential in terms of aesthetics. I don't think most of us really gave much thought about what a lens flare was before that.
Michael Wakelam: 0:26
Hey, welcome to this week's episode of the Creators Society Animation Podcast. I'm Michael Wakelam. Our podcast focuses on conversations with creators from all across the industry, discussing how they got their start and chatting through their career. I'm a writer and creator based in London and the Creators Society is based in LA and exists to support those in the animation industry. We've had a diverse range of guests, but today we have our first compositor. You may not be a compositor. I'm not although I do dabble. But most likely your work has been touched by a compositor at some point. A couple of weeks ago, Peter Demarest posted a show reel on LinkedIn which really blew me away. So I dropped him a note and the next thing you know, a podcast episode is born. Take the time to take a look at Peter's compositing show reel. There's a link in the shownotes. I love this chat. We talked about Peters career and some specific films and turning points in compositing and also how the role has changed and is changing with virtual production on projects like The Book of Boba Fett and The Mandalorian. Peter recently started at Pixar. So we also touched on how the role of compositing is changing in animated features. Regardless of whether you think you're interested in compositing or not I encourage you to dig into this episode because I always find that I learn the most from episodes I least expect to, not just here but in other podcasts as well. Before we jump in though, if I can ask a favour if you enjoyed this episode, please share it with colleagues or friends are on social media. And if you use the Apple podcast app, please rate and comment. Thanks so much. Without further ado, Peter Demarest. Hey, Peter, thanks for joining me today.
Peter Demarest: 1:58
Nice to be here. Thanks for having me, Michael.
Michael Wakelam: 2:00
Now, you know, we have never met before this call and no offence, but I didn't even know who you were until a couple of weeks back when you posted your reel online. And we all see a lot of reels and clips as we scroll through LinkedIn. But Shahid Malik had commented on it. And we both know Shahid, great guy. But his comment said something like, 'Damn all my favourite shots in one place.' So that made me watch it. And it not only kind of blew me away, obviously, the content was amazing, but I guess it was also pretty cool realising that you were able to be a part of so many iconic shots over the years. Now you've spent a lot of- a large chunk of your career at ILM. And you've also had stints at Double Negative and WETA fx and other big VFX houses. Now you're at Pixar, which is super interesting, but we can't talk about that. And you've worked on too many films to mention in one podcast. They include obviously, Harry Potter, King Kong, Star Trek, Transformers, The Avengers, Jurassic World, the Star Wars-es including The Mandalorian, Book of Boba Fett, and even the Batman. But before we get into that, all of that, what I'd like to do is hit rewind and talk about how you got started down this creative path and then we'll get into compositing for all these great shots and compositing in general. So where did it all start for you?
Peter Demarest: 3:20
Well, like a lot of my colleagues, I think it started as a kid sitting in a dark theatre, you know, watching a giant spaceship overtake the screen. Although actually Star Wars did come out a little before I was born, my father was was a big fan of genre movies. So from I think, the earliest age I can remember, he would bring me to the theatre, and we'd sat there. Watched The Empire Strikes Back and ET, I think I was five when ET came out. But I, I probably saw it about 10 times in a theatre. So that was, that was setting the table. I think that was probably his fault. He also showed me movies like Terminator, and Aliens, at a much too young age. But there was probably less of a stigma than that there is now. And then, when these movies started playing on HBO, they'd often be preceded or succeeded by a show, I think it was called behind the magic, something like that, where they- a crew would go behind the scenes and talk to all the crafts people making these amazing films. It didn't dawn on me then but probably later, when the word 'job' entered my lexicon that you could actually get paid for doing that kind of work. I think at some point in high school we were tasked with giving a report on a renaissance, a modern renaissance person. For some reason, the first person that came to mind was George Lucas. So in the course of doing that report, I came across all these different disciplines in special effects and it just sounded like a really neat thing to do. So that was the kernel of it. I remember my dad's- for my dad's 50th we went and saw Jurassic Park which had just come out. It was a book that we read together. We came out of the theatre and we were just both gobsmacked. He actually bought a computer, not too long after that, and downloaded Photoshop. That was the beginning of it. I learned that programme inside and out. I think that was version three. I probably don't know it- if I opened it today, I probably wouldn't know it at all. I was dabbling in that in high school. In college, I read that Industrial Light & Magic had an internship programme. So I went to an art school, the Rhode Island School of Design, which had a film, animation and video division. So that was my major, but it was very independent minded and more fine arts inclined. So they'd graduated a couple of people who, who went on to become pretty famous, like Gus Van Zandt, and the cinematographer Robert Richardson. But apart from that, not so much, and they were more known for their animation than anything else. So when I told the head of the department that I was applying for an internship at ILM, he was not optimistic about it, he said, they don't really like us out here. And funnily enough, my roommate and I both applied for the same internship. Lo and behold, one day I got a phone call. Turns out I had been offered the internship. The summer before my senior year of college, I flew out to California from Rhode Island, interned at ILM in the training department. So that was my foot in the door.
Michael Wakelam: 6:39
That was a great start.
Peter Demarest: 6:40
The training department was a great place. So this was the summer of '99. So Star Wars: The Phantom Menace (Episode I) had just come out, back then you would watch these tiny little QuickTime movie players in the browser to catch up on the trailer. And you know, me and my, my college buddies must have watched that for hours, just scrutinising every frame of it. And then, so when I got to ILM, they were ramping down from having expanded quite a bit to get that movie done. It was a very quiet time. I think everyone was just exhausted. And a lot of training material had been generated throughout the course of the production of that film. So my first job was actually watching hours and hours of training material and organising it.
Michael Wakelam: 7:31
That's fascinating, because Star Wars: The Phantom Menace (Episode I) was really, you know, those prequels, for whatever you think of the prequels and I know people are divided on those, but they were so groundbreaking in the way that we make movies and VFX and leaping forward another massive step weren't they.
Peter Demarest: 7:47
Absolutely. Just the sheer volume of work too. I mean to consider that that many shots were done under one roof is pretty astounding today. So I had no idea what these people were talking about. But after watching hours and hours of it, I think some of it started to sink in. You know, at the time, there were some clear divisions in the labour. There's animation and lighting, compositing. And I didn't really understand what these disciplines were what they were about. But at that time, I think there was a clear indication that you really needed a computer science background, or experience programming to get into lighting, or at least that's what we were told. And you had to understand some math to get into compositing. So none of these were appealing to me, because I didn't know any coding. And I was pretty awful at math. It was frustrating because I knew where I wanted to be. I just didn't know what I wanted to do. I couldn't figure that part out. By the end of the internship, I had met a few people, made really good friends with my managers in the training department, and was hopeful that maybe when I graduated college, I would end up back there.
Michael Wakelam: 8:58
Did that influence what you did in your senior year and what you focused on?
Peter Demarest: 9:02
It did. Fight Club had just come out and that was a really influential movie amongst college seniors. Everybody wanted make their version of that movie. So I had all these terrible ideas for incorporating visual effects into my senior thesis project. But at the time we were all shooting on film and the first digital video cameras had come out. So people like Danny Boyle was making 28 Days Later and Steven Soderbergh was experimenting with the format, but I think our department only had two and they were already claimed by the documentarians so there was no way I could do that. And getting film onto the computer at the time, was still a very expensive prospect for a student. So I threw all my ambitions to make a visual effects masterpiece in college, which is probably good because I could only animate an asteroid.
Michael Wakelam: 9:58
It's interesting though when you look at, you know, just 20 years ago, but now what kids who are at school have access to it's really incredible. Anyone can make a VFX film, I guess, of varying qualities, but the things you see coming out of college now are incredible.
Peter Demarest: 10:16
Absolutely. If you look at the student entries that the VES Awards, I mean, they put a lot of professional visual effects to shame, the quality and the skill, the attention to detail is, it's really intimidating. I'm glad I'm not a student right now.
Michael Wakelam: 10:33
So after your senior year of college, what did you do after you graduated?
Peter Demarest: 10:37
My girlfriend at the time got a job out in the Bay Area, I didn't have a job offer. But we moved out to San Francisco, sort of on a whim and I was confident I could get something, a bit overconfident. There were a few jobs open at ILM, in dust busting, which I'm not even sure is its own job anymore. My old manager in the training department had switched departments and was in information systems. And she said, If all else fails, I can probably get you a job. I didn't get any other job. So I knocked on her door again. And my first job at ILM out of school was as a basically an admin, I was her assistant. So it was completely unrelated to visual effects. But the department was writing all of the production software, so everything that the production department used to manage the pipeline and keep track of assets going through. So that was a really interesting entry, sort of through the back door, because our principal job was to go through the entire company and learn the pipeline from A to Z in as much detail as possible.
Michael Wakelam: 11:52
Wow, what a training ground for a newbie that is trying to figure out you know, which kind of job in which department to go in. That's incredible.
Peter Demarest: 12:01
So this was 2000, and was right before the .com crash. So there are all kinds of characters floating around at the time, but they hired a consultant, who I think was formerly at Bell Labs. And he came out and the enterprise was to basically just design an entire process flow of how a bit gets created at ILM and how it ends up on the screen. So I was sort of his trail he was not very socially adept. So I was in a sense, this translator between engineering and normal person speak.
Michael Wakelam: 12:42
So I guess from there, you somehow went on to roto and paint I think a lot of people in VFX start their career in roto. Was that an easy transition for you going through that kind of, into that art based- it's kind of an art based job, but it's very laborious.
Peter Demarest: 12:57
Sure is. At the time ILM had a wonderful training department, which we spoke about. So after work, I would jump on a free machine and start training and roto seemed like the most obvious place since it was almost entirely tedium and it was a skill I could practice. And it didn't involve a whole lot of math. So yeah, I spent a lot of hours after work, learning how to roto. But at some point, it became apparent that I wasn't going to be able to get a job at ILM in roto anytime soon, so I had to jump ship. About a year later, they were making The Matrix sequals a friend I'd made it ILM was the head of the roto team over there. It didn't end up getting a job on The Matrix sequels, but they, the company ESC, was starting work on some other projects. So he got me a job as a roto artist on Catwoman.
Michael Wakelam: 13:56
The famous cat woman.
Peter Demarest: 13:58
I wish I could say it holds up. I mean, some of the work that that they incorporated into that film was really groundbreaking. Maybe not the roto work, per se, but that was my first first artist job in the industry.
Michael Wakelam: 14:10
And where was that? Did you? Did you move for that job? Or?
Peter Demarest: 14:13
No, fortunately, I was living in Oakland at the time. The company ESC Entertainment which was started by Kim Libreri who's done pretty amazing things over at Epic in the time since. They were in Alameda on a naval base. So they had been set up there to produce The Matrix sequels and had started working on some Warner Brothers projects.
Michael Wakelam: 14:34
I guess, you know, from there, you had to jump into comp at some point because you've got so many amazing films that you've worked on it. Was it a natural progression for you to to jump into comp and at at that point. What kind of software were you using was, was that pre Nuke?
Peter Demarest: 14:49
That was pre Nuke. So at the time, let's see when I first started learning, compositing, each company had its own solution. So there was no universal software that everybody was using at that point. But then studios started adopting Shake, which was a node based compositor, which was a bit new and really confusing to anybody who had been training in After Effects or layer based packages. ESC was, was the first time I had been exposed to Nuke. And we were really busy on that project. But every free hour I had, I would just browse the network for neat shots and jump in to the comp scripts after hours and and see what I could pull apart. Always remembering to save a copy and not writing over anyone else's files. There weren't always protections against that. I had to teach myself a lot. It's pretty easy to do when you had amazing artists to work from. So I just stole and faked it until I made it so to speak.
Michael Wakelam: 15:52
Yeah. And Shake was the precursor to Nuke and Fusion all these node based tools I guess. And it was a pretty strong tool until Apple bought it and decided to shut it down.
Peter Demarest: 16:06
Famously, yeah. I think just about every compositer or still has a copy of Shake lying somewhere. The back their bookshelf. Yeah. ESC was Shake based. And from there, I well, they shut down actually, while I was there, but they were- they had several projects lined up. And then I think things just fell through. And that was the first in a long line of visual effects companies I seem to have left in my wake.
Michael Wakelam: 16:32
Well, I think there's more of a an issue with the VFX model business model, isn't it then, than anything else, and the small margins and chasing tax breaks around the world?
Peter Demarest: 16:42
That was probably the beginning of that, that race to the bottom. That was disappointing. People loved that company. They were doing some amazing work. They had a huge R&D department. It was really a loss when they shut down.
Michael Wakelam: 16:54
And so when they shut down, where did you end up?
Peter Demarest: 16:57
There was a smaller company called The Orphanage. They were headquartered in the Presidio in San Francisco. They had been started by some ex ILM people, there was a small Mac division called the rebel Mac unit, which had a pretty accomplished history working on the Star Wars prequels. And a few of them decided to splinter off and start their own thing. So I think it started as a company that was, that marketed itself to indie filmmakers to do visual effects for indie films. But they quickly found out they couldn't make any money that way. So they pivoted to, to studio projects. They were small but scrappy, and they worked on a lot of high profile films.
Michael Wakelam: 17:38
Yeah, that was Superman- Superman Returns, I think they did.
Peter Demarest: 17:42
Yeah, that was my first job as a compositor. I'll be honest, that was so long ago. I'm not even sure how I made the jump over. But I worked there in roto before, so they knew I would put in the hours. And they were using After Effects quite famously. They were a devoted Mac house at first. They are one of the few people doing visual effects for major studio projects using After Effects
Michael Wakelam: 18:06
That blows me away.
Peter Demarest: 18:07
They did some amazing things with it, particularly in the Iron Man movies. They designed all those interior, HUD graphics for the helmet, and I mean, talk about influential, it really changed the way HUD graphics were done.
Michael Wakelam: 18:21
Definitely. And I think that's the type of thing that you'd expect to be done more, I guess in After Effects. You went back to ILM at some point.
Peter Demarest: 18:30
I did. I started back as a roto artist because that was the only job I could get on the second Pirates of the Caribbean sequel (Pirates of the Caribbean: At World's End). I tried to take on some comp light shots when I was there, just so I could get some visibility. You know, at that time, I think now it's less common, people entering the industry can jump right into compositing, but at that time in the early 2000s, it was still very difficult to just jump right in to comp. You had to do some time at a more entry level job like roto or something else. I'm not sure why that is I was I was coming in from the
Michael Wakelam: 19:09
I think that's still fairly common to see outside but... graduates go in as roto artists and then transition. It's kind of a, almost like a, pay your dues kind of thing and I don't know if that's the same at every studio, but I do hear a lot of those stories that people are starting on I guess even their lower paid jobs as well and you know, something that graduates just want to get their foot in the door just as you got your foot in the door as an admin you know, at ILM, anything to get your foot in the door and I've heard stories of people starting as a receptionist, slash roto artist, you know. You know anything to get that, that credit and, you know, move on up from there. Well, let's fast forward a few years because you started obviously as a compositor and looking at where you are now you're very successful at that and you got the hang of it. You know, you worked at various places on various films and I think you travelled to WETA for a while. You were at DNEG for a while. And then you really settled in back at Lucasfilm or ILM for a chunk of years working on some great films as a lead, compositing lead. Can you talk about, you know, some of your highlights there at ILM because we can look at your reel and go wow, this is incredible. You know, you know what it was some of the highlights of the films? I mean, you mentioned that you you'd seen Jurassic Park with your dad and then getting to work with Jurassic World would have been incredible. I'm sure.
Peter Demarest: 20:37
The Pirates of the Caribbean films were pretty massive. And I don't remember the exact number of shots ILM did but the quality of them at the time. So my first job as a compositor at ILM was the third Pirates of the Caribbean movie
(Pirates of the Caribbean: 20:52
At World's End), and they developed the Davy Jones character for the second, but seeing him, seeing the renders come off the farm. And just the quality of the surfaces and the animation was mind blowing. You know, the at The Orphanage compositing was taking the raw renders, and then trying to really doing a lot of work on them to make them fit the scene. And that was down to issues of scheduling and budgeting, the artists there were incredible. But seeing the renders of ilm, for Davy Jones and all those other characters was just mind blowing, you really didn't have to touch them at all. One of the benefits of working at a place like ILM as a compositor is, in some ways, there's less you have to do, you can focus less on integrating a rendered character into the scene, because most of that work has already been done for you. You know, these are people who not only invented the technology, but invented all of the methodologies since, to make things more realistic and emulate all the physics of light. So what we could do as compositors is really focused on what everyone talks about as being the last 5% of a shot that makes it perfect. But yeah, that was an amazing project. Star Trek was also, gosh that was an incredible film and really influential in terms of aesthetics. The I don't think most of us really gave much thought about what a lens flare was before that film.
Michael Wakelam: 22:24
Yeah, famous lens flares, there are a lot in that film. I'm guessing that came down to your department.
Peter Demarest: 22:30
That's right, that was part of the aesthetic directive for the film. Yeah, there was a person on set with a flashlight shining it into the lens to generate in camera flares. But it was such a necessary part of the aesthetic that we knew we had to incorporate that into any all CG shot, or, you know, or blend in to any plate photography too. At the time the tools were a little more rudimentary than than where they're at now. I mean, now we have dedicated software for lens flare. Some of the leads on that project spent a lot of time developing different players to match what was coming through in the plate photography. But you know, we're 13 years out from that film and, and they're still shooting lens flares into the camera for all the Star Trek projects they're making now.
Michael Wakelam: 23:17
Well, and I guess we're 13 years out from that film, and we're still using the Optical Flares, plug-in in After Effects to recreate what happened back then, you know, that was, that was quite amazing when that came out.
Peter Demarest: 23:29
That's true. I can't remember the first time I used that. I've gotten to know that very, very well. But yeah, from I think I worked on the first Transformers movie for a little bit. I wasn't a staff artist at ILM. At that point, I was a contractor. One of the reasons for getting into visual effects is really with Lord of the Rings, you know, with WETA coming to prominence in New Zealand, you start to realise that there are opportunities around the world to do this kind of work. And that was really exciting. My wife and I both had international passports at the time, we really just wanted to get out there. And you know, we both lived in the States for most of our lives and just seemed like a great opportunity to get out and see more of the world. So prior, you know, earlier while I was a roto artists we'd lived for a year in New Zealand but after Star Trek, it felt like the right time to make another jump. Double Negative had done some incredible work on a number of projects like the Batman films and Children of Men, which was probably my favourite movie since Blade Runner came out. I contacted them and got an interview and they offered me a job. So my wife and I packed up and then moved to London, contract for a year. So we figured we'd start there and I started at Double Negative
Michael Wakelam: 24:44
Yeah, and Double Negative really kind of made a name for themselves. I guess a bunch of studios made a name for themselves with the Harry Potter films, starting out over here and all the tax credits and wanting to take advantage of that and local production and that really grew the VFX industry. in London in the UK, but then yeah, as you say went on to be doing so- such great work on so many films.
Peter Demarest: 25:07
It's hard to find the film these days that double negative hasn't worked on. Probably be hard pressed to find one they haven't contributed to. But yeah, they they've done incredible work, obviously.
Michael Wakelam: 25:19
And obviously not- no longer just in London, they're around the world because they need to be around the world because there's tax breaks in various countries. So you spent a year was it just a year in London?
Peter Demarest: 25:32
It was just a year. Yeah. It's a little expensive over there.
Michael Wakelam: 25:35
Yeah. You're telling me. Bay Area, you're in the Bay Area. That's not too good!
Peter Demarest: 25:39
But the funny thing was, we came home from London and '$12 cocktails?! That's incredible.' I don't think you could buy a pint of ale for $12 while we were living in London,
Michael Wakelam: 25:53
Did you go straight back to San Francisco? Or did you go to New Zealand from there?
Peter Demarest: 25:57
No. I had gone to New Zealand for a short stint on Avatar, right before we moved. I realise this is a little out of order. But that's kind of a funny story that the move to London was probably about seven or eight months in the making. So we just started packing up all our stuff. And I was working at another studio that ended up closing down Image Movers, the short lived Robert Zemeckis enterprise. I had gotten a short contract to finish up with A Christmas Carol there. WETA had actually contacted that studio to see if they had anybody rolling off that project who would want to come work on Avatar. No one really knew what Avatar was. But everybody knew that it was James Cameron's next movie that they shot in stereo that it was probably going to be big. And I just about given up hope of ever working on it. They put my name forward Image Movers put my name forward. I talked to the recruiting staff at WETA, I had been there for a year on King Kong. So they they had a file on me and I guess I did okay, they offered me a job to come back. So I said yes, immediately. I'm not sure I discussed that with my wife. Yeah, I left her in a bit of a lurch. And she had to finish up all the packing while I ran off and chased my visual effects dreams. So I owed her big time for that one. But yeah, at the time, they were WETA was still working in shake, but they had developed a deep compositing pipeline. So that was probably the biggest advance in compositing technology wise that I'd ever seen. You know, they had these enormous scenes and CG generated environments. And the way normally renders would come off the farm is you'd have your character renders. And this is just kind of a general approach anywhere but you you'd render in layers, just in case, let's say you needed to make a change in your character. Your background was okay, your CG jungle was okay, but the character needed some additional animation. The scene would be broken up into layers, so that you can render out the character multiple times and just leave the background alone. But in order to marry those two renders together, the typical approach was to render holdouts. So what that means is just taking the character and using it as a matte object in the jungle render. So when you get your jungle render it would, you'd see all your plants, but there would be a black hole where the character would be and then we would marry those two together in the comp. The problem was, every time you would make a change to the foreground character, if you're rendering with holdouts, you would have to re-render the background anyway. And this would just be prohibitively expensive. There was some adoption of a deep shadow technology, or deep shadow renders, which were Pixar animation. And I'm not even sure that Pixar was using it in this way at the time. But WETA was probably the first company to develop a deep comp approach where you would have, you'd have pixels that were not just aware of the silhouette of a character, but they would also store information about where that character existed in the 3d scene. So instead of having that holdout rendered into the background render, now you just have the background render without any holdouts whatsoever. Using the deep information you could combine your background jungle render, and your foreground, alien render, and just marry them together without any artefacts. They had built this system on top of Shake. It was early, but this was huge. I mean, there's there's no way they would have been able to finish that film on time without it. But that changed everything from a compositing perspective.
Michael Wakelam: 29:49
Yeah, definitely. I want to jump forward a little bit to I guess a little bit- a lot to The Mandalorian because you worked on The Mandalorian. And like you said that deep compositing was is a massive leap forward in comp work. When you get to The Mandalorian, and virtual cinematography and shooting in the volume, as they call it, you've got all of these CG elements on the screen, where you're not putting them in, in post anymore. You are doing some of that. But you've also got miniatures and stop motion, and you've got actors, and actual, you know, set elements in that volume. That was also one of the biggest leaps forward, wasn't it? In changes I guess, in your workflow. So I'd be interested to hear about that from a compositing perspective.
Peter Demarest: 30:36
It's, it's interesting, it, it harkens back to one of the earliest techniques for putting actors in a different environment. There was rear projection and front projection, like in 2001, the prologue was all front projection. So the concept is very similar. I mean, it's almost the same. But the difference is that this time, the background isn't canned, it's not shot ahead of time, it's, it's live on the set. You had a lot of what were typically green screen setups, where you'd have a character in front of a process screen, green, or blue. These are pretty straightforward bread and butter compositing shots at this point, but there, there are always problems inherent with shooting that way. You know, there's there's spill lighting from the screen onto the foreground character, the DP is lighting something to an imaginary background that they can't visualise. So the lighting on the character, the photographed character doesn't always quite match with the background that goes behind it. So these are all things that this method of shooting on an LED volume seeks to solve. And in a more cost efficient way. The immediate conclusion that was drawn was, oh, this is gonna, this is going to completely eliminate compositing, which it doesn't. It created this, this methodology of shooting created a whole discipline of visual effects that didn't exist before. It's been a boon to the industry that way. I worked on the the first season of The Mandalorian, I was the compositing supervisor on the project for ILM. So there was just a lot of figuring out of how compositing comes into the workflow. It involves a bit of branching out to be sure, but there's still things that you want to happen on in that background, that are just too expensive to render at the time, you know, things like fire steam, you know, any kind of particle driven effects, we're still using, canned elements that can be triggered on stage. It's the same with a lot of the games industry, you know, in gaming, a lot of the particle effects are generated ahead of time and pre rendered, but triggered within the game. So we took a similar approach, we're probably not that far off for where things will be simulated in real time. I mean, who knows? It's probably happening right now as we speak. But there's also a lot of quality control and compositing that goes into the elements that appear on screen.
Michael Wakelam: 33:15
Yeah, so you've got a lot of that post production is been brought forward to pre production, and projected onto the screen and- not projected- shown on the screen, then filmed and then touched up again in compositing in post. And, you know, I've seen some of the clips, and we did a virtual production shoot last year. And there was some things that we just could not get ready in time for the screen, they just went to a quality that we wanted. And so we did the same thing that we had seen The Mandalorian do, in putting a green rectangle behind the actor, so you still get the lighting from the screens on to the characters in that way. But you can still do a regular post process afterwards.
Peter Demarest: 33:55
And the benefit of compositing elements, well, you're heavily reliant upon roto, to be sure. Let's say for whatever reason, the background is not performance. Maybe there's some artefacts that creep in, you know, something happens where you need to replace the background. And you're set up for success in a number of ways. When you shoot this way. You have all the incident light falling on the actors, so they're much better integrated into the scene.
Michael Wakelam: 34:22
Especially when they've got highly reflective suits on.
Peter Demarest: 34:25
Yeah, that helps. But you will need roto to separate anything from the foreground off the screen, but you also have the entire background available. It was assembled for the actual shoot. So you have that scene and you have the tracked camera on the day. So all that information is already there for you. So it's not quite that easy, but it's more or less someone hitting the render button. You get your backgrounds you have the camera already set, you really just have to focus on the fine details like the the edge characteristics and things like that, that makes it much easier to composite when it comes through in post.
Michael Wakelam: 35:05
Yeah, it's interesting to see you know where that will go, as you say, when you can have fire and water and smoke in real time, it's going to, it's going to take it another step forward.
Peter Demarest: 35:16
But it's also created a pathway for compositors to have jobs on set. Because there is an element of live compositing that happens when you have the practical set that's built in front of the camera. And then you have the LED screen behind them. It's never a perfect marriage of colour temperatures right off the bat. Any LED volume setup is going to have live controls for somebody to come in and balance colour. That is a pretty amazing thing to be doing, you know, just playing with dials and matching colour, right in front of you in real time. And, you know, let's say there's a virtual prop that looked great in the concept art that's blue, maybe, but maybe on the day, someone wants to make it green. Well, it's pretty easy to change right there and then. You know, being attuned to, to colour relationships, and, you know, the integration of foreground and background elements. It's a natural fit for, for compositors to, to step into these onset roles.
Michael Wakelam: 36:21
Yeah, yeah, that's really interesting. All right, I'd like to just finish up by chatting a little bit about comping animation, you know, as we were talking about before. I guess our podcast has been predominantly talking with people in animated films, animated TV shows. But you know, large parts of these VFX films are completely animated, because until Tom Cruise does it, then we don't actually shoot in space. So- and in animated films, historically, there hasn't been a separate comp department, we talked about that. Often, the lighters would do all the compositing of their own shots. And in VFX, compositing has been used to make the fantastical look real. But you know, we're seeing more and more compositing come into the animated feature pipeline because they've got these not real realistic looking films, but really stylized looking films like Spider-Man: Into the Spiderverse and The Mitchells vs the Machines and The Bad Guys. And there's a lot of compositing of 2d elements. And I think if
you look at Spider-Man: 37:23
Into the Spiderverse, or maybe it was the Mitchell's the credits, there was like 150 compositors, which is unheard of for an animated film. So it's really interested in and obviously when we're not going to talk about Pixar, but just your views on the industry. And is that making it more attractive for people from the VFX industry to kind of make that sideways transition into animated projects?
Peter Demarest: 37:46
Well, that's a pretty big can of worms to open.
Michael Wakelam: 37:51
Let's just finish off with that. Yeah.
Peter Demarest: 37:55
Well, compositing it has typically been the realm of, of cheating, really, so and what I mean by that is things that were just too expensive, time consuming, or either took a lot of processing or R&D to develop in 3d can be cheated in compositing. In a traditional visual effects pipeline and, you know, in a lot of cases, with photo real animated films, what compositing can do really well and quickly is it can emulate characteristics of a real camera. So you have, you know, lens vignetting, around the edges of a frame, chromatic aberration, specular highlight, blooming, things like that. And these are all subtle touches that give you that sense that something's been photographed by a camera. So in big visual effects projects, where you have a lot of all CG shots, this is the realm of the compositor to take something that's very crisp and pristine, and just dirty it up and introduce some some chaos and randomness to it. And it's also been- compositing is sort of the neutral workspace where different render packages get integrated together into a unified scene. So if you have a character being rendered from RenderMan, maybe some effects being rendered from Houdini, they all meet in the comp. While those different 3d packages can have an awareness of the different pieces, you know, through, let's say, VDPs, were an effect can light up a character, something like that. There's always a bit of massaging that needs to happen to integrate the elements, the disparate elements into a unified image. So that's always been the domain of compositing. And depending on your animation pipeline, you know, you could be rendering everything out in one render, or you could be rendering using multiple renders. There's always a utility to have compositing in that respect. But what's interesting now and I think what really blew the doors off was Spider-man: Into the Spiderverse, is you have these stylistic components where we're getting away from that sense of photorealism. You know that sense of it a unified lighting into much more stylistic and expressive modes, compositing is, you know, that's, that's sort of the natural progression from 2d animation. That's where all your 2d bag of tricks exists now. So I honestly don't know how spiderverse is made. And I kind of don't want to know, because it's just- the ultimate effect is so mind blowing, you have what's clearly sort of a base 3d render, but you know, all of these 2d textures coming in and out of time. And it's an interesting time right now, because a lot of people make these films are coming from the visual effects background, where there's always been this marriage of 3d renders, and 2D filtering of those renders. And now you're starting to see something in the animation realm that's just more experimental and more expressive. And it's a really exciting time. We're both going down that road of photorealistic animation where things have a very physical dimensionality. I mean, look at
the new Avatar: 41:10
The Way of Water films, those will be spectacular. You have, you know, things like Arcane or Into the Spiderverse, which are, you know, those are not photo real by any means. They're using every trick in the book to achieve, you know, something that you just can't do in either render or compositing alone. It really opens the door for different expressions of story. And I think it's a very exciting time. And what's great is that there's more content being- I hate to use the word content, but it's- there are more stories being told and you know, all these different avenues theatrical, it's streaming, and they're creating more opportunities for people to get into the industry through multiple different doors.
Michael Wakelam: 41:58
You know, I guess we'll see where all of that goes. As far as the industry converging, you know, we've got film and games converging and film and animation VFX and animation converging. So there's, there's a lot of crossover, whereas I think years ago, there used to be very siloed, and everyone was in their own industry. So it's going to be really interesting to see where it all goes.
Peter Demarest: 42:20
And the good thing is, everybody has more time to watch all this stuff, right.
Michael Wakelam: 42:26
Yeah, we wish. Well, it's been so great to chat today. I really appreciate your time, Peter.
Peter Demarest: 42:31
Thank you for having me, Michael. It's been a great conversation.
Michael Wakelam: 42:36
Thanks for joining me, I hope you enjoyed that conversation. If you'd like to get in touch or shoot us any feedback, please email podcast@thecreatorssociety.org. You can find me on LinkedIn and other socials. As mentioned at the top, please subscribe, like or share the podcast. Special thanks to Rich Dickerson for the music Mike Rocha for the mix and our exec producer Eric Miller. Thanks again. See you next time.
