Latest / Bad Decisions Podcast / #71 The TRUTH about AI & Filmmaking (Nobody Tells You) with Jim Geduldick
Transcript
- 0:00Virtual production isn't just one thing.
- 0:01It's not just LED walls, it's live events, it's broadcast,
- 0:05it's for feature films, it's for episodic, or it could be for V
- 0:09tubers and Tik Tokers, The Weather Channel in real time.
- 0:12That's virtual production. People may not have known it,
- 0:15but those things that have been around for a while, that is
- 0:17virtual production. This whole conversation of like,
- 0:20well, do you be a generalist today or a specialist?
- 0:22Like I look for Swiss Army knives.
- 0:24They don't have to be an expert in everything from if you know
- 0:28the basis of the core fundamentals, animation,
- 0:30filmmaking, of storytelling, and you start to learn the
- 0:33technology, the technology is always going to change on you.
- 0:35But if you have those core fundamentals, you'll be able to
- 0:38take that throughout your entire career.
- 0:39How insane is that? There is a person who can do
- 0:42this alone in his house and be offered $5,000,000 and then
- 0:45there's other people complaining that there's no opportunities.
- 0:48People forget just how powerful the stuff is in our pocket or
- 0:51our backpacks. Like you have a studio in your
- 0:53backpack. So now, especially with AI,
- 0:56Pandora's box has been opened. There's no excuses.
- 0:58There's no reason you can't go out and make the thing you see
- 1:02in your mind. We're here.
- 1:09Welcome to the podcast. Thank you guys.
- 1:11This is. Amazing.
- 1:12This is amazing. This is first of many.
- 1:16We have Jim for the first time in person in Vancouver.
- 1:20That's right, we did this last time, two years ago.
- 1:23It's it seems like it wasn't that long ago.
- 1:26Yeah, man, it's insane. We got to do it.
- 1:28I think it was like one of our it was the 1st 15.
- 1:31In the 1st 15 episode that we did at BAD, it was called Bad
- 1:34decision Coffee break that time and we used to drink coffee in
- 1:37the beginning. Now we have our coffee here.
- 1:39All right. Cheers.
- 1:41Cheers, I hope you guys watching are also drinking some sort of
- 1:44coffee because this is going to be 1 hell of a conversation.
- 1:48We just talked about the things that we're going to cover.
- 1:50Of course, we got the the man to talk to when it comes to virtual
- 1:54production. So going to cover that.
- 1:56We're going to talk about AI. Everyone wants to hear about
- 1:58that. In the room.
- 2:00Yes, we have to and at the same time sigraph is happening.
- 2:03We literally just stole. Jim from Sigraph.
- 2:06We promised to bring him back in two hours.
- 2:08Yes, he was in the middle of a talk.
- 2:09We just stole him and said you're coming to the podcast.
- 2:12Yeah. Thanks.
- 2:14For having me. Of course, the next time we do
- 2:16this, you're going to be a 40 Gaussian splat, right?
- 2:18We just discussed this. Yes, we're going to do it
- 2:21virtually. Like very virtually.
- 2:23Yes, yes, but it's going to be like with next level tech, but
- 2:26this time physically some. I'm really glad to have you on
- 2:28the pod. Man, yeah, thanks.
- 2:30It's good to be in person. Yes, then virtual.
- 2:33But yeah, it seems like we did. That podcast didn't seem like 2
- 2:38years ago. Yeah, time flies.
- 2:39Time flies, man. How is Vancouver, by the way?
- 2:42I love this city. It's, it's a lot of fun.
- 2:45This, you know, this conference is, it's got a lot of history
- 2:49behind it on, you know, it's been talking about and showing
- 2:53advancements of computer graphics for a couple decades
- 2:57now. You know, it's like there's
- 2:58people that have been coming here, you know, for, for since
- 3:03the first one, you know, like over 30 years ago.
- 3:06Wow. So yeah.
- 3:07And this and this is a show that's really interesting
- 3:11because it's at the crosshairs. If it's not just visual effects
- 3:14here, it's not just computer graphics, it's not just
- 3:16animation, it's not just AI these days.
- 3:20And it's a real mix of like where things are today in in a
- 3:25lot of different facets of computer graphics.
- 3:27And you know, like we say, virtual production AI is a big,
- 3:31a big hot topic of, you know, generative machine learning,
- 3:33computer vision. But like all the big studios are
- 3:36here, all the researchers are here, independent studios,
- 3:42doesn't matter kind of what part of the industry if you're
- 3:44creating pixels or manipulating them in some way.
- 3:48This is one of the better shows throughout the year that that
- 3:53happens. And it's it's cool to be back in
- 3:56Vancouver because you know, shows like this, like on real
- 4:00fast FMX, things like that are centric around like, you know,
- 4:04animation and visual facts and computer graphics, they hop
- 4:08around. So like, you know, the the show
- 4:10will be back in LA next year for 2026.
- 4:14And it's cool because you get a lot of global people coming
- 4:18together and you get to run into old friends, you meet new
- 4:22people, a lot of interesting conversations at, you know,
- 4:26these are nerdy talks that we get into about what what is the
- 4:32state of art today in whatever field?
- 4:36Because, you know, this is covering games, it's covering,
- 4:40you know, visual facts. It's covering very, very deep
- 4:43research on to the future of a lot of the stuff that we all use
- 4:47today in virtual production and games and traditional production
- 4:51and visual facts. And it's always good to see kind
- 4:55of like what's just on the horizon coming at shows like
- 4:58SIGGRAPH. So it's really cool to be here
- 5:01and, and this city, you guys know, you lived in this city.
- 5:04This city is a lot of fun and it's in in Vancouver is it's
- 5:09also a hotbed for animation, visual effects, motion capture,
- 5:13virtual production. So there's a lot of lot of lot
- 5:16of cool things going on in this city.
- 5:18A lot of people travelled from Europe.
- 5:19We were just blender meet up. A lot of people from Sweden,
- 5:23Ireland, Finland. We were shocked.
- 5:25So many people travel all the way from Europe to join this
- 5:28conference here. I got to give it to Farhat I
- 5:30wasn't going to make it here, forced me.
- 5:33He was like, we got to do cigarettes.
- 5:34In Vancouver, how many times it comes in like we should come?
- 5:37Here and I want to say thank you for bringing us because you see,
- 5:42as you mentioned, when you come here, not only do you realize
- 5:45what's on the horizon, you're going to meet other people who
- 5:48are like minded, all the nerds that are just like us.
- 5:51And I feel like at least for us, for me and Farrah, because most
- 5:53of the stuff that we do is in the studio behind the computer,
- 5:57you can feel a little bit lonely at times.
- 5:58Even though we're hanging out together, we don't get to talk
- 6:01to everybody else in the industry.
- 6:03So what I want to recommend to everyone watching and listening
- 6:05if they've never been to anything like SIGGRAPH or Unreal
- 6:07Fest, if you get the chance, fucking do it because it's so
- 6:11fun to meet random strangers who are doing something similar to
- 6:15you. And then you get to realize
- 6:17there's actually a lot of people in the industry that you don't
- 6:19see. You just see comments online.
- 6:21And The thing is, sometimes comments, I mean, I'm sure we're
- 6:24all aware in the comments section, people rage, there's
- 6:26negativity. But when you actually go to
- 6:28these conferences, it's all positive.
- 6:30You actually see real people and you talk to them.
- 6:33Sometimes you might meet your future partner.
- 6:35Yeah, not in relationships, I'm talking.
- 6:38About maybe you never know. SIGGRAPH just started.
- 6:42What are you up to it? What are you coming to?
- 6:44SIGGRAPH. I came for him to do the
- 6:45podcast. That's the reason you're here.
- 6:48All of this. Wait, wait.
- 6:49What's going on here, guys? But yeah, I think it's it's a
- 6:52really good place to come and meet people and just hang out.
- 6:56Yeah, and there's, I mean, you guys saw it today at the Blender
- 6:59meet up. There's there's meetups that are
- 7:01just community driven and you know, open source like blender
- 7:06is and like some other tools out there bring together like you'll
- 7:10run into really deep tech engineers, you'll run into
- 7:14animators, you'll run into CE OS of studios and a lot of it is
- 7:19all is all passion driven. Like obviously a lot of people
- 7:22are in the business, but at the end of the day, you get to see a
- 7:26lot of like minded people and the connections.
- 7:29I mean, that's the whole thing that's fun about filmmaking and
- 7:32all this stuff is like collaborating and people.
- 7:36It's like, you know, we didn't meet in person until today, even
- 7:40though we've been talking for the past couple years and and
- 7:44then you kind of never know where stuff can go.
- 7:46It's like you wind up working together.
- 7:48You could hire that person, you can get hired by that person.
- 7:51You can just collaborate on that or a support call and you'd be
- 7:55like, hey, I'm stuck with this meta human problem.
- 7:57What do I do? You know, so it it and then
- 8:00that's what you're getting to is like building your network of
- 8:03community. That's the biggest thing because
- 8:05as you grow in your career, you're the sum of your, your
- 8:10Rolodex, your connection. So like when you're looking for
- 8:15the person to help fill out a team for a project or when
- 8:18you're looking for support or you're just looking to bounce
- 8:22ideas off of people, you're always kind of looking into your
- 8:25friends and you know, your, your wider area connection to nobody
- 8:29knows everything, of course, you know, So it's like, who did, who
- 8:32do you reach out to? Yeah, I have the perfect example
- 8:35of that. So when we were working on the
- 8:36Ultimate Unreal Engine course, we were looking for character
- 8:39designers because we were making the summarize, and we ended up
- 8:42working with really talented artists.
- 8:43But just now at the convention, we randomly started talking to
- 8:46people, and then we got each other's Instagram.
- 8:49I saw some incredible character artists for the next project.
- 8:53I know exactly who to call. Yeah.
- 8:54So if you're online, if you DM that person, they may never see
- 8:57it. It might.
- 8:58Now you met in person. Met in person.
- 9:00Exactly. So now if I need characters, I
- 9:02know exactly who to call. If I need animators, I know
- 9:05exactly who to call. So the connections that you
- 9:07build is incredible. It's so important to network and
- 9:10also get each other social media because sometimes, for example,
- 9:13your ex. I love it because you always
- 9:15post updates. The new things that are coming
- 9:18out, the new things that you find out when I'm looking at X,
- 9:20when you are posting something, I know that, for example, a new
- 9:23AI model is out or a new technology is out or a new
- 9:26update is out. If you are not following the
- 9:28right person, you might not even get the right.
- 9:30Information. Yeah, yeah, there's.
- 9:32And there's a lot of information flowing these days.
- 9:36So it's like it even for me, it's, it's hard to be on top of
- 9:40everything, especially with all the new tech that's been coming
- 9:44out and updates, even on real engine updates or nuke updates
- 9:47or Blender updates, there's always something new that you
- 9:53don't catch because you know you're not on social media 24/7
- 9:56or hopefully not. And there's like, I'll look for
- 10:02your guys post, you know, I'll, I'll, I'll look at somebody from
- 10:06the epic team, you know, that's working on a particular thing.
- 10:09And I always know our our buddy Matt Workman, you know, to see
- 10:12what. Different.
- 10:13People are up to and you know, information is coming at us so
- 10:19fast that you're like, well, you could miss this new generative
- 10:22model that dropped or you could miss this new fab plug in that
- 10:26got released on, you know, on the on the marketplace.
- 10:29So having that to kind of know where things are or, or like
- 10:34going out to like 80 level or any of those industry sites
- 10:37where you're just like, their job is just to get updates and
- 10:40find cool stuff. It's a good repository to find
- 10:43these different outlets that are covering that.
- 10:45So like with you guys doing the podcast and doing the Unreal
- 10:49Engine courses, becoming more of those resources for the
- 10:53community, then you wind up. Yeah.
- 10:55I mean, you've seen it how your growth has gone and since when
- 10:58we first talked to now and then you guys putting out the
- 11:00courses. It's it, it becomes a place
- 11:04where people go to find you out and you never know, you wind up
- 11:08getting work that way or you just wind up meeting new people
- 11:11that way. So it's really cool to to kind
- 11:13of build that type of community. Learning 3D VFX and animation
- 11:17should not take you years. That's exactly why we designed
- 11:21the Ultimate Unreal Engine program, which is our best work
- 11:24yet. This is our masterpiece, which
- 11:26is going to take you step by step from beginner to pro in
- 11:29Unreal Engine in just 14 days. And by the end of it, we'll get
- 11:33to create a beautiful cinematic short film just like this.
- 11:37We'll take you step by step through pre production planning
- 11:40out your scenes to setting up your environments using
- 11:43photorealistic assets really quickly.
- 11:46We will teach you modeling, texturing and animation.
- 11:49We'll even teach you how to use meta humans to create beautiful
- 11:52digital characters that look real.
- 11:55We will teach you AI workflows to speed up your process.
- 11:58We'll teach you how to set up your lights so that your scenes
- 12:01look cinematic. We'll teach you how to use
- 12:03cameras inside Unreal Engine to set up beautiful shots.
- 12:06Animate the cameras not just digitally but using virtual
- 12:10production techniques like using your phone to move the camera in
- 12:14the scene, which is super. Fun.
- 12:15We will teach you motion design tools so you can create an
- 12:18amazing title sequence at the end.
- 12:20We'll teach you simulations as well, like cloth destruction and
- 12:24particles, and you learn all of this in 14.
- 12:27Days and for those of you who are listening to this podcast,
- 12:30we have a very special price. Use the code podcast in the link
- 12:35below this video and you will get 300 $100 off from our
- 12:39Ultimate Unreal Engine program. And to make it a no brainer for
- 12:42you, if you hate the course and you hate our teaching style,
- 12:46we'll give you your money back because we've never had a single
- 12:49person say that before. You can even go to our Discord
- 12:51community and check with our members what they think about
- 12:54the course. And by the way, Speaking of the
- 12:56Discord community, we have 24/7 support for all of our students,
- 13:01which means if you ever have any issues while learning, you can
- 13:04always contact us. We'll respond, get it solved.
- 13:08Even if the problem persists, you can send us your Unreal
- 13:11Engine files and we will fix it for you.
- 13:13Send it back to your Never stuck.
- 13:15And the reason why we're so proud of this 24/7 support is
- 13:19because most courses don't actually offer that.
- 13:21So if you're learning and you get stuck, it's over and you're
- 13:24lost. So farad.
- 13:27Click on the link below this video and use the code podcast.
- 13:31PODCAST, in case you didn't know, let's get back in.
- 13:36So let's not forget why we're here today, right?
- 13:39We wanted to talk to you for a very long time because the last
- 13:43time we had a conversation, it was about virtual production.
- 13:45I would say a lot of the topics that we covered and a lot of
- 13:49time, I know it's been 2 years, it went by fast, but so much
- 13:52that's a. Lifetime.
- 13:53Yeah, exactly. So much has happened.
- 13:55There's been lots of new updates.
- 13:56And as you mentioned, since we've gotten more people in the
- 13:58audience, I'm sure there's so many people who want to know
- 14:01about virtual production. They might not have a clue as to
- 14:04what it is. Some of them may have a little
- 14:06bit of an idea, but they want to get into virtual production.
- 14:09They don't know how. So I guess the best way to sort
- 14:12of set the conversation is to understand what is virtual
- 14:16production. Which is a little bit of a
- 14:18loaded question because it's many things.
- 14:22And, you know, right before COVID and going into COVID,
- 14:25virtual production was, was kind of marketed and, and put out
- 14:32there as, as the savior solution, especially during
- 14:34COVID, because we couldn't, we couldn't go out and shoot
- 14:36together because of COVID restrictions.
- 14:40And virtual production isn't just one thing.
- 14:42It's not just LED walls, It's not just one real time engine or
- 14:46one, you know, media server. And it's, it's a bunch of
- 14:52different technologies, it's a bunch of different disciplines
- 14:55right under this moniker that is virtual production.
- 14:58It, it got a lot of popularity, obviously over COVID because of
- 15:04we couldn't get out and get together and shoot.
- 15:06And we had these, you know, bubbles and we had these
- 15:08restrictions. It was a maturity of technology
- 15:12that started catching up to where things were.
- 15:14So obviously things like Unreal Engine had the capabilities as
- 15:20as kind of the Swiss Army knife that allowed you to do things of
- 15:23being able to generate pixels and create virtual art
- 15:27department assets that then could be displayed and projected
- 15:32onto LED volumes. You could use projectors.
- 15:35We're talking about, you know, broadcast technology mixed with
- 15:39animation mixed with motion capture and facial capture
- 15:43technology and all of these kind of modalities of virtual
- 15:46production are under the umbrella of that.
- 15:50So it's not it was deemed or marketed.
- 15:53A lot of people sold it as like well, virtual production is it's
- 15:56LED volumes mixed with Unreal Engine.
- 15:59There is a truth to that. A good chunk of that work does
- 16:02happen via those things, but it also hits a different industry,
- 16:07right? Sectors like it's live events,
- 16:10it's broadcast, it's for feature films, it's for episodic, it's
- 16:15for just motion capture that you may either post solve into a AAA
- 16:21game or even an indie game, or it could be for, you know, V
- 16:25tubers and Tik Tokers. Like virtual production doesn't
- 16:29have to be one thing. It's many things to many people
- 16:32and it's really on like how you're using, but you're using
- 16:36different technology to get to an output that normally would
- 16:41have been more traditional. What we think of, you know,
- 16:44normally we would have had to capture assets, then we would
- 16:47have had to go through a post production visual facts and CG
- 16:50pipeline. And then you do a post render
- 16:52and then that's what you get. That's your, that's your output,
- 16:55you know, a composite pipeline. Now that we start to get into
- 16:59more of the real time maturity with real time engines being out
- 17:03there, real time compositing things that we've seen over
- 17:07there of like in Unreal Engine, we've got composure for kind of
- 17:11like mixed reality and real time compositing.
- 17:13We've got, you know, different engines with hardware, like, you
- 17:18know, black magic with Ultimat for real time keying.
- 17:21You have lots of software that can do real time keying in this.
- 17:23So it's, it's really matured in terms of what you can do under
- 17:31that kind of moniker of virtual production.
- 17:34And it's really based off of, you know, what, what are you,
- 17:37what are you, what story are you trying to tell?
- 17:39What are you trying to achieve technically with the tools?
- 17:42Because virtual production can mean something totally different
- 17:45to somebody that's just doing motion capture for an output to
- 17:50a video game or for, you know, for, for live streaming that
- 17:54data. Or you could be working in
- 17:56camera visual effects where you're working on feature films
- 17:59and episodic and and live broadcast where you're whole
- 18:04system may be like 1 style of working way.
- 18:06Like virtual production been around for The Weather Channel
- 18:10in real time key. That's virtual production.
- 18:12Yes, I mean, and, and people may not have known it, but like
- 18:16those things that have been around for a while that that is
- 18:19virtual production. You know, you're using different
- 18:22tools for in output where you're using different technologies of,
- 18:27of visual effects as it's matured or mixed reality as it's
- 18:30matured. There are aspects of VR that are
- 18:33virtual production, you know, so it's, it's, it's a blanket term
- 18:37that covers so much. So, you know, it's, it's really
- 18:42comes down to like when, when we say, you know, the, IT allows us
- 18:45to work virtually, you know, like you guys can be in Dubai, I
- 18:49can be in LA, we're, we are working virtually.
- 18:52You know, it's like you can do things with Zoom that are
- 18:55virtual production, you know, so it's really a, a, a subset of
- 18:59tools and technologies and it really gets kind of tweaked
- 19:06based off of what your workflow is.
- 19:08What I hear is that there is lots of opportunities that if
- 19:13you have built your foundation. So if you know your foundation,
- 19:16if you know the tools that you you, you have to know, then you
- 19:19can select and decide, OK, I want to go into game
- 19:23development. I want to go to live broadcast,
- 19:25I want to go to field man animation.
- 19:27As long as I build the foundation, then I can be
- 19:30specialized into going into any of the direction that I want.
- 19:34Is that true? Yeah.
- 19:35And, and there's this whole conversation of like, well, do
- 19:38you be a generalist today or a specialist?
- 19:40Like, you know, it used to be in visual effects and animation and
- 19:44the early days of motion capture and virtual production, there
- 19:47were a lot more specialists because there, you know, these
- 19:50things that, you know, Jim Cameron and Bob Zemeckis and,
- 19:56you know, people like Glendarry and Robbie Derry and the team of
- 19:58techno props. Early on or creating the first V
- 20:02camps and and creating and using the first like early days of
- 20:05motion builder, you know, and and Mobu as everybody knows it
- 20:09for so long is you know, a lot of this stuff was getting
- 20:13developed out of necessity because these tools didn't exist
- 20:16for film makers and storytellers.
- 20:18So you had. Motion Builder by the way, I I
- 20:20didn't use that before. So Motion Builder is an Autodesk
- 20:25product and it was one of the tools early on, probably built
- 20:30out of the early days for Avatar of being able to record motion
- 20:36capture data, deal with animation data and that they're
- 20:40obviously Maya had been around and pretty early as days of the
- 20:43Alias Wavefront and and early things like that.
- 20:46But motion Builder of being able to bring in animation data,
- 20:50capture motion capture data, use it to then bring it in and
- 20:54retarget it, whether you're in, you know, Max Maya and other
- 20:58tools, a lot of tools back then, like over 20 years ago, some of
- 21:02those don't exist anymore. Motion Builder, a lot of people
- 21:05know other different, different tools for animation retargeting.
- 21:10There's obviously a lot of custom tools at different
- 21:12studios too that get specialized for that.
- 21:15But in terms of, you know, what we're saying about, you know,
- 21:19the basis of like starting off of like generalist and, and
- 21:23specialist is, you know, a lot of people today that do get
- 21:28hired or who I look forward to hire out for different teams and
- 21:30different projects is like, you know, I look for Swiss Army
- 21:33knives, just like the tools are Swiss Army knives looking for
- 21:37people that have a diversity. They don't have to be an expert
- 21:40in everything from Unreal Engine to motion builder to knowing how
- 21:43to work with motion control and LED volumes.
- 21:45There still are specialists. You're still going to search out
- 21:48an an LED engineer or an LED tech who knows everything that's
- 21:52going on on, you know, the wiring to dealing with the
- 21:55processors to dealing with and being able to swap out power,
- 21:58power supplies and you know an LED panel.
- 22:00You're still going to have specialists there just like you
- 22:04would with dealing with motion capture, facial capture, rigors,
- 22:08you know, modelers, all of that. But if you have a good breath of
- 22:14the the basics, you know, you know a little bit about
- 22:17photography and cinematography for shot composition.
- 22:20You know, it's like if you know the basis of the core
- 22:23fundamentals of animation, of filmmaking, of storytelling, and
- 22:27you start to learn the technology, the technology is
- 22:29always going to change on you. But if you have those core
- 22:32fundamentals, you'll be able to take that throughout your entire
- 22:34career. Learning about art and learning
- 22:38about different disciplines is always going to benefit you as
- 22:42you progress. And whether you go more into the
- 22:44creative art or you go more into the technology, if you're
- 22:48writing the tools and you're creating the tools for all of us
- 22:51as end users, you're going to make better tools for all of us
- 22:54because you understand the creative side too.
- 22:56And sometimes there's been a gap in that where you've had
- 23:01developers or engineers that haven't been on set before,
- 23:05haven't actually maybe used the tool in production.
- 23:08And it's, and it's great to have companies who are developing the
- 23:11tools come on set with us or come see how we're using, you
- 23:15know, meta human animator and how we're using, you know,
- 23:19Blender and Maya and, and Touchdesigner and all these
- 23:23different things out there. Because at the end of the day,
- 23:26the frustrations that we go through are either a gap in
- 23:30knowledge of how to do thing, you know, do something A to B in
- 23:35one tool or the other, or that hasn't been solved yet.
- 23:39And either you're going to solve it.
- 23:41Somebody's out there, you know, writing a blueprint class right
- 23:44now to put out on fab to go, hey, well, retargeting used to
- 23:49be much harder than it was, you know, and in any animation
- 23:53package, doesn't matter if it's Unreal or Maya or Blender or Max
- 23:57or pick your flavor of DCC. But if you can take those core
- 24:02fundamentals, you know, of whichever style of the industry
- 24:08working on, especially in the creative field, like you can
- 24:11take that. So it doesn't matter if you're
- 24:12creating graphics, real time graphics for Tomorrowland when
- 24:16you're doing something for dead mouse or you know, it's like you
- 24:20can replace the tool, they get replaced and and sometimes they
- 24:24go away. So, you know, the big thing I
- 24:27tell a lot of people who are asking about like either
- 24:30pivoting in their career or starting out was like, don't get
- 24:33married to anyone tool because it'll get changed.
- 24:37Like you're married to the, the process of like storytelling and
- 24:41filmmaking and the basis of that.
- 24:42But like, you know, don't get married to one specific tool
- 24:46because either that tool goes away, it changes.
- 24:49You're going to learn something else, so just be prepared for
- 24:54change like it will happen. If you're enjoying this
- 24:57conversation, we just wanted to remind you that we come back to
- 25:00you every single Tuesday with another episode of the Bad
- 25:03Decisions Podcast. It will be available on YouTube,
- 25:06Spotify and Apple Podcast. Now back to the conversation.
- 25:10100% I feel like that's one of the things that we talk about in
- 25:13the podcast as well you. Always say that actually.
- 25:15We use Unreal Engine mostly because we believe at this
- 25:19moment in time in 2025, it is in our opinion one of the most
- 25:23versatile software's because of the amount of outputs you can
- 25:26have. Anywhere we travel to, we see
- 25:28Unreal Engine artists doing crazy things, right?
- 25:31We talk about game devs. We have Tomorrowland visuals
- 25:34being done with Unreal Engine. We've got people in the
- 25:37automotive industry using Unreal Engine, real estate using Unreal
- 25:40Engine, but that could change, right?
- 25:43Maybe three years from now. It's a different software.
- 25:44And as you mentioned, if you got your fundamentals down, if you
- 25:47know lighting, if you know composition, camera work, when
- 25:51you transfer that even into AI filmmaking, let's say in three
- 25:54years time, you're making films with Runway ML because that is
- 25:56what everyone is doing. If you know how to set up your
- 25:58compositions, you can tell the AI how to do it, right.
- 26:01So as you mentioned, I think if you have your fundamentals down,
- 26:03that is the most important thing.
- 26:05But I want to ask you something about what you said you look for
- 26:08in people when you guys are hiring.
- 26:10Yes, the generalist specialist, that's something we should
- 26:12cover, especially when it comes to AI.
- 26:13I'm going to get to that. But what are the things that
- 26:16you're actually looking for when you guys are hiring, when
- 26:18yourself or other people in your team are hiring?
- 26:20Imagine if Pharaohs is submitting a job application.
- 26:22I'm. Looking for a job right now.
- 26:24What should I have in my tool bag to get you guys's attention?
- 26:28Guess what, the first thing is not technical at all.
- 26:32It's, it's, I will look at like your people skills, your soft
- 26:35skills, how you communicate within a team, how you play
- 26:39within a team, you know. How do you know that in an
- 26:42interview before the guys even? It's a little bit of, it's well,
- 26:46a lot of getting to know people from like a dry hire, meaning
- 26:50like somebody just submits a resume or a reel or anything
- 26:53like that. Usually it's like I said, back
- 26:55to your network. I find out about other artists
- 26:58either I, you know, see work on social media.
- 27:02I get to work with them in a Southern way.
- 27:04You know, it could be from like, you guys mean like, hey, I, you
- 27:07know, I found this artist. They're really good at this and
- 27:09it's word of mouth. And that's another thing about
- 27:11putting yourself out there is like, you know, you come across
- 27:16these people and like I've gotten hired from putting work
- 27:20out on social media and and over my career too.
- 27:23And so like putting yourself out there, but like having those
- 27:25communication, those soft skills is a really big part is like a
- 27:29lot of what we do, whether you're on, you know, working on
- 27:33AAA games, you're working on feature films, episodic, even
- 27:35from an indie level and an indie studio, like you are made or
- 27:39broken by the team and your support group that you're
- 27:42surrounded with. So that's why I say like
- 27:45technical skills are really great.
- 27:47Artistic skills are really great.
- 27:49If you can't communicate in a team, it's going to be really
- 27:54hard. Like you're either not going to
- 27:55vibe with the team or you're going to have friction with it.
- 27:58So like, either you have to learn how to do that because
- 28:01like, not everybody's going to be, you know, all rosy and
- 28:04cheery and like, and some people that are really, really
- 28:08talented, they're introverts, like they do really well by
- 28:11putting on their headphones and they just lock in.
- 28:13I do the same thing. It's like sometimes you just
- 28:16want to go tunnel vision because you know, you either have this
- 28:19vision of something you want to execute on or you just need to
- 28:23disconnect from the hive and and go it alone for a little bit.
- 28:27But then you've got to come back into it.
- 28:29It's like you can only get so far by yourself and then you
- 28:34know your support group of like finding answers to problems you
- 28:37don't know. And even though there's, you
- 28:39know, there's cloud and ChatGPT and all these models out there
- 28:41now that can help you answer it, you know, you still like the fun
- 28:46part is, like I said, it's, it's collaborative.
- 28:48It's like getting to meet people in person.
- 28:50It's getting to like find out about the people because you
- 28:53don't know what that's going to bring for good or bad into that
- 28:57project or into that story. And yes, the technical and the
- 29:03creative and the art skills are, are definitely parts that I look
- 29:07at too. But it's just like, how am I
- 29:09going to vibe with you on a project?
- 29:12And if we don't get along, you could be the best you can know
- 29:16on real engine backwards and forwards.
- 29:19If you're Dick, I'm not going to want to work with you.
- 29:22So it's really like, so it's really why I say like that's
- 29:25really important. And if you're look, there's a
- 29:28lot of developers, there's a lot of talented artists, there's a
- 29:31lot of musicians, you know, there's a lot of painters out in
- 29:34just in general, they're not outwardly like, you know, and
- 29:39they might be really timid and you never know, you might be
- 29:41able to break that person out of their shell and really help them
- 29:45or you just don't know. And that's the thing why I say
- 29:48like those skills are really important.
- 29:50It's not just the technical skills.
- 29:52Those skills can be learned. You can learn any application
- 29:57out there, but the communication skills, the, the foundational
- 30:03skills that you should be learning.
- 30:05It's why a lot of people look at like, well, do I go to school
- 30:07for this or do I just take a like a course or do I just go to
- 30:11like a seminar and I say like explore all of them?
- 30:15You just, you never know as people learn differently.
- 30:17Yeah. You know, like I'm a visual
- 30:20learner, Like, like I could take a bad decisions course and go, I
- 30:25just like, I know this part. I just want to skip ahead to
- 30:27module 5 because like that's that one thing I need to learn.
- 30:31And that's why YouTube and social media is great because
- 30:33there's a lot of answers out there to be like, I know how to
- 30:36do these parts. I'm just looking for that one
- 30:38little thing. Is it this one thing with like
- 30:41how to set up metahuman, you know, cloth the right way.
- 30:45Am I going to Chloe and am I going to Houdini?
- 30:47Like when I'm what am I doing here to do a cloth SIM?
- 30:50And the cool thing about it is like there's so many resources
- 30:53for us out there, but coming back to it, it's like, you know,
- 30:58those skills that we're all talking about here, you know,
- 31:02and, and anybody should be in that mode of like always
- 31:07learning, you know, just like there is nobody that knows
- 31:10everything. If somebody says they know
- 31:11everything, I would say that they're full of it, you know,
- 31:14and there are experts in in whatever fields and everybody
- 31:18kind of goes to them. You know, we have the people
- 31:20that we all follow to be like, oh, that, that person just
- 31:24they're amazing at Houdini and they do really, really amazing
- 31:28work. How are they doing it 'cause
- 31:30it's always, it was always that thing of like you saw this
- 31:32amazing shot on YouTube on a film or whatever and it was
- 31:35always like, what camera did you use?
- 31:37And it's like, well, that wasn't the answer.
- 31:38Just 'cause you have you get access to a RED camera or an FX3
- 31:42or, you know, a Panavision system.
- 31:44It like, it doesn't mean you're going to be a Roger Deakins or
- 31:47or, you know. You're not going to get the
- 31:49yeah. Yeah.
- 31:50And that's the other thing too, is like so many of the tools
- 31:53that we're talking about are very accessible today.
- 31:55Like previously they were behind a paywall or they were custom
- 32:01tools that you only got access to if you worked at a studio.
- 32:05So now, you know, especially with AI, Pandora's box has been
- 32:10opened of like, there's no excuses.
- 32:12There's no reason you can't go out, go out and make the thing
- 32:17that you have you see in your mind.
- 32:19And and like you were saying is like, yeah, well, OK, a lot of
- 32:22us, there's a lot of favoritism here too.
- 32:23Unreal Engine for obvious reasons, like is it perfect?
- 32:28No, no, no. There is no DCC.
- 32:30There is no application that is 100% perfect.
- 32:34There are definitely ones that are like we say, you know the
- 32:38Swiss army knives that allow you to do a lot of different things.
- 32:41You can do cinematics, you can do motion capture, you can do
- 32:44games, you can do feature films, and there's maybe other tools
- 32:50out there that get to it, but they may not be real time.
- 32:54And that's a lot of things we're talking about here is like the
- 32:56advancements in real time for cinematics, for feature films,
- 33:00for virtual production, or like maybe you just want to make
- 33:03short form TikTok videos, but you don't want to put yourself
- 33:05out there. So you create a a AV tuber
- 33:08version of yourself and you know, people can share because
- 33:13they're like they feel really timid of sharing their face on
- 33:16on camera. So they might start off as you
- 33:19know, using an avatar that they created in a real time engine or
- 33:23ADCC. And all of a sudden they go from
- 33:27the regular nine to five job that they're not happy with.
- 33:29They find a new means of making a living, and they're much
- 33:34happier than sitting behind a desk during TPS reports.
- 33:37Oh my. God.
- 33:38All of this were not possible a decade ago.
- 33:41Now we have all most of these softwares are free now.
- 33:44You can get education on YouTube for free, you can switch your
- 33:47career, you can change whatever you're doing in a matter of
- 33:50months. And this a decade ago was not
- 33:53possible. And now this is not only limited
- 33:56to North America. People are doing it in different
- 33:58continent, different parts of the world, still sharing their
- 34:00work and still getting jobs and still like changing their
- 34:03careers. It's crazy how much
- 34:06democratization is there now with all these tools that are
- 34:09available. The problem I have is we see it
- 34:12in our comments section all the time, in our Discord community
- 34:15all the time. People say that there's no
- 34:17opportunities, there's no jobs, there's no money to be made.
- 34:20And that is true for some people.
- 34:22But at the same time, you see people who are making hundreds
- 34:25of thousands of dollars, millions of dollars alone.
- 34:28We were just talking about this Corey Williams, who has a
- 34:32YouTube channel called Just for Kids.
- 34:33I want everyone to look this up. He made a kids show on his own
- 34:39solo called Silly Crocodile and he uses a controller because he
- 34:45uses Unreal Engine, so it's real time.
- 34:46So he connected Silly Crocodile to a controller and he's moving
- 34:50the character using the controller while his iPhone is
- 34:53in front of him, recording his face.
- 34:54So he gets to animate the character at the same time.
- 34:58And he does that show, I think once a week.
- 35:00So once a week he has an upload. He's got about 800,000
- 35:03subscribers. We got 200,000, he's got
- 35:06800,000. And he just offered him
- 35:08$5,000,000 for the IP of Silly Crocodile and he refused it and
- 35:14he refused it. And we get to talk about that.
- 35:16But how insane is that? There is a person who can do
- 35:18this alone in his house and be offered $5,000,000.
- 35:22And then there's other people complaining that there's no
- 35:24opportunities. Yeah.
- 35:26And and and look when we look at global production at scale, like
- 35:31yeah, there there are different demographics of to like what
- 35:35people have accessible to them from technology to all that.
- 35:39Yeah, there's definitely people that have less and that, you
- 35:45know, even are you? Talking money wise or?
- 35:47Just accessibility to the end and it could be because of of a
- 35:51monetary thing. It could be for that.
- 35:53But there I think if you're passionate enough and you have
- 35:58the drive, you won't take no for an answer.
- 36:01You'll you'll find a way to get to that thing.
- 36:06And we've seen a lot of success stories of it.
- 36:08And it's like, you know, look, this, this being an artist
- 36:13creating technology, whatever it is, it's not for everyone.
- 36:17And it is hard work. Like if you want to work on
- 36:21feature films or you want to make your own game or you, you
- 36:24want to put out things like what Corey does, what's like
- 36:27crocodile. It's like you can see how
- 36:30passionate Corey is when he puts his vlogs up.
- 36:33And the other Corey, Corey Strasberger, who does blue like
- 36:37another great example of somebody when I met Corey for
- 36:41the first time, it on real fast. I was just like, oh, how big is
- 36:43your team? How do you and it's like 1 dude,
- 36:46one guy. That's fucking.
- 36:47Insane with like teaching himself motion capture and
- 36:52that's really cool about that and and yes like there's still
- 36:55parts of the industry with virtual production and things
- 36:57like that that are still expensive.
- 37:00So some of the stuff is is still not as democratized.
- 37:03We're getting there with markerless and and and.
- 37:06Helping with that. It is, yeah.
- 37:08And there's always two sides of the coin of like, what?
- 37:10What's helping? Because right now we have that
- 37:12whole question of like, people pivoting in their careers and
- 37:15asking like, well, should I get into visual effects?
- 37:19Should I get into film making? Should I get into the game
- 37:21industry? Because they hear about
- 37:23rightfully so, all this volatility of like, oh, this
- 37:26studio closed down or this film got shelved or like, you know,
- 37:29people are chasing subsidies all over the globe.
- 37:31What do we do? But it does leave options open
- 37:36to to figure it out. Like, you know, this is arguably
- 37:40the most powerful computer and camera in the world.
- 37:46What you could do with this, even from a virtual production
- 37:48standpoint is like, and you guys talked, we've all talked about
- 37:51this, an iPhone or an iPad hooked up to even a, a game
- 37:57ready PC or a game ready laptop. You you have your virtual
- 38:01production. Studio you have the V cap.
- 38:03Let's talk about it. You can use your iPhone to scan
- 38:06any environment. Put it in Unreal Engine.
- 38:08You can use it to animate the face of any fucking character
- 38:12because meta human creator now native inside Unreal Engine, you
- 38:15can bring in any character. It was so freaking difficult
- 38:18when we did our. First tutorial.
- 38:20But now it's so simple, you can get dinosaurs in there, you can
- 38:23get zombies in there, can animate the face in a few
- 38:25seconds. Actually you don't even need
- 38:27that. You can use your voice now to
- 38:29animate the face. You can connect, like I said
- 38:32with Corey, a controller and control the character.
- 38:34There's so much you can do. As a virtual camera, handheld
- 38:37camera inside your scene, do whatever you want.
- 38:39VCAM yeah, it's tons. It's and and that's what's cool
- 38:43about the the kind of technology side is like giving also giving
- 38:46yourself enough room to like experiment and explore.
- 38:50And I know I'm like, if that's not your day-to-day job and you
- 38:53know, that feel very privileged that I get to play with a lot of
- 38:56tech and, and problem solve for, for a lot of different things.
- 39:00But the end of the day too, it's, you know, I'm giving
- 39:02myself time to kind of look at what the next pieces of
- 39:07technology. And sometimes they're not in
- 39:10Hollywood at all. They're not in broadcast.
- 39:12Or that it could be technology that is, you know, more sports
- 39:17science related, or it might be from construction and AEC to be
- 39:22like, well, those guys and gals have been using Lidar and site
- 39:25scanning tools for forever. We use those tools.
- 39:29That's a big thing too, is like look outside your swim lane and,
- 39:33and you know, pay attention. We have so many tools to find
- 39:37things that are outside our, you know, our peripheral vision in
- 39:40terms of like usability of technology.
- 39:42It's like, you know, it's like an artist, if they only stay in
- 39:46their town and they look around painting the same things, like
- 39:48they're going to just keep generating the same things.
- 39:51But like the artist that travels and takes in a lot more of a 2
- 39:55is going to, you know, you're going to have a lot more color
- 39:59to your output. You're going to have a lot of
- 40:01different scenes. It's the same thing of like, you
- 40:03know, all of this technology that gets thrown at us day in,
- 40:07day out. It's it's hard to stay on top
- 40:09of, but it's like, you know, really, what do you want to do
- 40:13with this stuff? Like do you want, what type of
- 40:15storyteller do you want to be? You're either going to be a
- 40:17storyteller to sell something. To someone or you're gonna build
- 40:21a world and and sell the idea of like, come, you know, come
- 40:25explore this world and the mediums change like and you guys
- 40:29know this going to a lot of different immersive experiences
- 40:33or events or things like that. You've got things like Cosm and
- 40:36the Sphere and you've got the pop ups that happen at like all
- 40:39the festivals, right? You know, Tomorrowland and all
- 40:42that. They're all using these bits and
- 40:44pieces of technology from virtual production and mixed
- 40:47reality and scanning. And obviously the target is
- 40:51like, all of this stuff is as real time as possible.
- 40:55Yes, nobody wants to sit around and wait for a render anymore.
- 40:58Not an. E-mail.
- 41:00We're we're becoming impatient day by day.
- 41:02Well, yeah, that that's the whole economy of having things
- 41:07instantaneous to us is like it, it, it creates the like, I want
- 41:11it faster, yes, you know. I mean, right now, if you have
- 41:14any questions about anything you ask ChatGPT 5, you get it in a
- 41:18millisecond. You're not going to wait for a
- 41:20render for three days. You know what I mean?
- 41:22We're and you're scrolling on TikTok and Instagram and you're
- 41:25getting fed videos every millisecond as well.
- 41:28That's why waiting for renders is becoming increasingly more
- 41:31challenging. Hence why I see the interest why
- 41:33people are using either Unreal Engine or blenders.
- 41:36EV now is getting a lot better as well, so people are moving
- 41:38into real time a lot more. There's there's still there's
- 41:41still a gap in some areas of this for real time tools on
- 41:45fidelity, right where you still may go to, you know, render man,
- 41:50you may go to, you know, vantage patch.
- 41:52Tracing engines. Yeah, and it and it's really
- 41:56like what your target is like are you going more photo real
- 41:58than you are stylistic? You know, there's this too is
- 42:03also speeding up because of machine learning, computer
- 42:06vision and generative is like we're starting to see some of
- 42:09this come into the renderers, both real time renderers and
- 42:12post renderers. So you know, just utility things
- 42:16of like upscaling getting faster of like, you know, most of our
- 42:21target stuff is like people aren't putting 4K renders on
- 42:24mobile Destin stuff. YouTube, though, is like your
- 42:29target is like you want to be able to put out at minimum base
- 42:32like UHD 4K output because of quality, because where are most
- 42:36people besides mobile watching? They're watching through their
- 42:39Apple TV on that. And I think it was just like
- 42:42last week where YouTube surpassed, you know, some of the
- 42:45streamers and some of the, the foundational, you know, outlets
- 42:50in, in terms of that and it, and it's like, where are you
- 42:54watching your content versus where you're watching?
- 42:57Like film is still different. And there is a there is a
- 43:00difference between content and film 100% there, there, there is
- 43:04a curation and there is a different process to what
- 43:08content is and how's that, how that is consumed and what kind
- 43:11of film and cinema is. And there's still a tier and a
- 43:15higher level to what that is. It doesn't negate all the up and
- 43:20coming Youtubers that are now creating their own production
- 43:24companies, which are then getting film deals.
- 43:26And we look at like Mr. Beast, perfect example, like what Jimmy
- 43:29and that team is doing. And they're little, you know,
- 43:32factory out there of like, you know, getting Beast games on
- 43:36Amazon. Amazon.
- 43:36Yeah, is the, and this goes all back to like putting yourself
- 43:42out there, getting discovered. You don't know when the
- 43:47algorithm is going to hit the lottery for you and some
- 43:50director CEOCMO brand discovers your stuff and is like, I like
- 43:57that. Hopefully it's I like it and
- 43:59they work with you other than taking your idea and just trying
- 44:01to do themselves. Because that happens, they will
- 44:03work with you because idea is very hard to steal.
- 44:06So if you are telling a story, it's it's in grading your DNA.
- 44:10So people, people want Mr. Beast to run that show, to run that
- 44:14competition. People want that specific
- 44:16creator to go and create that show for them.
- 44:19And I think big companies will also support that.
- 44:21As long as, as you said, you put yourself out there and it
- 44:24doesn't matter where you put yourself.
- 44:26If your channel is linked and if your channel is X, if your
- 44:28channel is Instagram. We had so many podcast guests
- 44:31that they said I posted something randomly and someone
- 44:35saw my work, hired me. We had people who said I wanted
- 44:38to work with D JS. I started making visual for that
- 44:41DJ without getting commissioned, without getting paid.
- 44:44I posted, they saw it, I was hired and I was commissioned
- 44:47that. We captured that's the old adage
- 44:49of two of like, spec work and like, do you take the risk, you
- 44:54know, in your early careers, like you've got nothing to do?
- 44:56Nothing to do is what you are just experimenting and learning,
- 44:59right? Even even even today, like doing
- 45:02like I still do spec stuff just to test things out.
- 45:04Like like doing there, there's a fine line of like if somebody's
- 45:08asking you to do spec work for free, that should be paid work.
- 45:11That's where you drove the money.
- 45:12The opposite. Then there's the like, well, is
- 45:15it worth the chance? Like if you know, would I take
- 45:19as a would I do kind of something with my own time?
- 45:22If you know, like, hey, you get a, a call from a friend or even
- 45:26a director who's just like, well, I got this idea.
- 45:28What do you think? Sometimes you just got a gut
- 45:30check and go, I'm taking a risk here.
- 45:33I'm going to put in my own time, my passion and my brain power
- 45:37behind figuring this out. And that's, that's an intuition
- 45:40thing you pick up over your career too.
- 45:41And just like back to the communication people skills of
- 45:44like being able to read people and get that vibe check and just
- 45:47be like this person trying to screw me over or they're really
- 45:50trying to just like work with me, partner with me and make
- 45:53something cool. So that's back to like you,
- 45:57you've learned this over time in your career.
- 46:00Like that, that intuition, the gut check the, the learning when
- 46:06you know to do things for free for friends, which I still do.
- 46:10Like if I get calls from certain friends that they've got a
- 46:13really cool idea and I'm like, I have the time and I then I'm
- 46:17like, I want to get in on that. Like I'll, I'll do that.
- 46:21It's not it for me too. It's not always about money at
- 46:25the end of it. It's really of like at that the
- 46:28heart of it, you know, I'm an artist like I want to create
- 46:31cool shit with friends. And sometimes the the good part
- 46:35of it is you get paid. And the other part too is like,
- 46:38it's a self fulfilling thing. Like, you know, and I think
- 46:41everybody has that, even, you know, for people watching or
- 46:44just like, well, I'm not an artist for like, I don't do this
- 46:46yet. And it's just like, well, you're
- 46:48going to find your way to whatever it is hopefully.
- 46:52And that's why like the whole kind of general specialist thing
- 46:55of like you got to try a bunch of different things out.
- 46:57Like you, you don't set off like somebody who starts off in roto
- 47:00and painting. They're just like they're not.
- 47:02Hopefully they don't want to do that forever because it's very
- 47:05tedious work. And it don't it will get
- 47:06replaced by machine learning computer vision at some point.
- 47:10Not today yet. It's getting there.
- 47:12But those people probably want to be animators or CG soups or
- 47:17VFX soups or directors or whatever it is.
- 47:19And you know, you're still paying your dues of learning by
- 47:23coming up like being a PAA Gopher and going to get like in
- 47:28the mix. You're always going to kind of
- 47:29like make your way throughout whatever industry you're going
- 47:34to do. And, and finding your path is
- 47:38always a constant re pivoting re you know, I've changed so much
- 47:44over my career. I've kind of stayed in, in, in a
- 47:47wider lane, but I've always known that being able to pivot
- 47:51into my career has been beneficial to be like, well,
- 47:55we're seeing these shifts and in filmmaking and visual effects
- 47:59and, you know, and CG as a whole, there have been so many
- 48:03pivot points over my career to just be like, well, if this door
- 48:08closes, you know, this one's going to open or you got to
- 48:11break down the door yourself and just like make your own way.
- 48:14So it's like a really big thing for, you know, those questions
- 48:17that you guys get through the community of like, where, where
- 48:20do I start? Where do I go?
- 48:22You know, it's a big thing too with obviously, you know, AI and
- 48:27where, where do creators, where do film makers go?
- 48:31If the promise of these tools that you know, is a bit of a
- 48:35hype mixed with, you know, practical application of where
- 48:40that leaves and it doesn't matter in the creative field,
- 48:44you're talking about any type of job being impacted by
- 48:48technology. So it's not just Hollywood, it's
- 48:51not just games, it's not just music videos or brand
- 48:53advertising or wherever it is globally.
- 48:56We're at a very big pivot point in this shifting technology.
- 49:01We haven't seen this ever and we haven't seen it in a in a long
- 49:06time and something that really changed all of us globally.
- 49:11So we go back to like being educated, understanding what's
- 49:14out there from these different things because you know how, how
- 49:19even from today, like a student today who's just trying to learn
- 49:22about math, like asking ChatGPT or Claude or anything.
- 49:25Like that same thing, right? The same thing with VFX and CG.
- 49:28Any industry is now getting this big hit from OK, what do I do if
- 49:33all of these things are getting automated?
- 49:35It's it's in every single industry now.
- 49:38Yeah, there's a there's look, there's a lot of fear mongering,
- 49:43especially in visual facts. People do have a right to be
- 49:48questioning or worried about that in certain types of roles.
- 49:53And that's why I say like, you know, playing with these tools,
- 49:57understanding what they can and can't do is really good.
- 49:59Obviously there's a cost to using these tools too.
- 50:02So is that viable? You know, like a lot of AI gets
- 50:08a Polish demo gets put out by AI company or, you know, generative
- 50:12company of choice. And they, they look really good.
- 50:16They're really alluring to a lot of people to be like, well, I
- 50:19don't have those skills. And if I prompt this, you know,
- 50:22or I or I converse with this AI, it'll just give me this output.
- 50:27There's a good and bad to that. It's like, yeah, you get speed
- 50:30and scale, but what are you giving up of that when you're
- 50:32not really learning anything, You're not learning the process.
- 50:35You're not learning good, bad artists struggle.
- 50:38And out of that struggle, a lot of times the happy accidents
- 50:41happen. Like, you know, you get a
- 50:45location gets shut down. You know, an actor gets sick.
- 50:48What do you do? You got to think on the fly.
- 50:50You know, you're not always going to be prompting like, hey,
- 50:53ChatGPT, my location just got, you know, screwed.
- 50:56What do I do now? You like you still have to use
- 51:00your brain power for a lot of this stuff.
- 51:02So it's, it's good to really be educated as what's out there.
- 51:07You know, obviously tools like this coming into, you know, our
- 51:11programs, our real time engines, our DCCS, our cameras like these
- 51:17are amazing advancements that do allow us to, you know, a, an
- 51:22artist with a vision and, and a basis of skill set and, and
- 51:28understanding of what story structure is and composition and
- 51:31photography and, and cinematography and lighting.
- 51:34Those artists that learn these new tools or these new
- 51:39platforms, those artists are going to get hired than other
- 51:43people that are just like totally against it.
- 51:45I'm not touching it. I saw a lot of this too on the
- 51:49switch over from film to digital, and there was a lot of
- 51:52people that are just like, Nope, that's sacrilege.
- 51:54We're only shooting film. Well, those DPS or those
- 51:57cinematographers or those people that didn't want to advance with
- 51:59it, they didn't get hired. Interesting, I see.
- 52:03OK. It's repeated the cycle.
- 52:05It's the. Cycle and it's the same thing
- 52:06too of like, you know typewriters to you know home
- 52:10computers and things like there's all these evolutionary
- 52:13steps and yes this one feels different because it is it is
- 52:17scaling so fast and you're seeing a lot of industries be
- 52:22affected especially in ours is it's funny that you you you see
- 52:28a lot of the platforms today especially generative based like
- 52:32they have a they have a target for Hollywood and.
- 52:35It's. Funny, it's a, it's funny
- 52:38because you know, it's like entertainment is an it's an easy
- 52:42target, you know, for, you know, pixel generation and output.
- 52:46But these tools are they're pattern recognition tools.
- 52:51They are good to do what we've collectively done up to today.
- 52:56At this point, it doesn't know what you're going to do
- 52:59tomorrow. It's not predictive.
- 53:01We don't have precogs like Minority Report that can see in
- 53:04the future, not yet. And it's good up to this point.
- 53:09But the, the, the I think the human element that still comes
- 53:12into it is like, even if Unreal Engine was totally generative
- 53:15and it was all faster than real time and all that, like if you
- 53:21could just let it go by itself and it just starts spinning out,
- 53:25you know, cinematics and it like, what's the fun in that?
- 53:27Yeah, like you, you, we still want to drive whatever that
- 53:32thing is, that tool, because we have a vision in our heads and
- 53:37the excitement of it is like this can help us get these ideas
- 53:41out faster. That's what was for me like 10
- 53:43years ago with Unreal Engine, was like coming from a
- 53:46background in cinematography and visual effects.
- 53:49I was like, I've been waiting for stuff to go real time.
- 53:53I've been dealing with post render work flows for so long
- 53:56and waiting for render passes and waiting for that e-mail from
- 53:59deadline to tell me that hey, these shots are done.
- 54:02Oh shit, wait, something messed up.
- 54:04I got to go back and re render. Let's take another three days to
- 54:07get out. You know, there's one sequence
- 54:10and and and this won't be the only technology shift like it
- 54:15will in our human history. We've always had these
- 54:17technology shifts. Now it's speeding up.
- 54:20So I think that's what's scaring people is how fast this is all
- 54:24moving. But if you take a step back,
- 54:28there's certain things that it is, it's displacing a lot of
- 54:31things, but it's not fully replacing a lot of things.
- 54:35How has it shifted the film industry over the past two years
- 54:38since we had our last conversation?
- 54:40Because AI has advanced a lot, we have different tools like
- 54:43Runway a left was just announced, which can generate a
- 54:47green screen for you or remove a green screen for you.
- 54:49We can turn a scene into a snowy scene, a dusty scene.
- 54:52It's pretty powerful stuff. Then Genie was announced as well
- 54:56by Google, where you can take an image and then turn it into an
- 55:00interactive world. It's very basic right now.
- 55:02You can go forward, backwards, left and right, but it is
- 55:05promising stuff. I want to know practical use
- 55:08cases from what you guys have seen.
- 55:10How has the film industry been using this stuff?
- 55:13There there's a gut check with all of these different platforms
- 55:15and and tools being put out into the market.
- 55:21The first thing is like you've got the two sides of it.
- 55:23You've got the the pay to play of like you're paying for
- 55:26subscriptions to get like render tokens of getting these
- 55:31generational outputs and then you have open source on both
- 55:35sides is a lot of these platforms on the pay to play
- 55:39have done a good amount of the heavy lifting.
- 55:41They have a lot of the tool sets in there versus going out, you
- 55:45know, building your own control net, training your local Laura
- 55:48using something like comfy UI flora Wii.
- 55:51A lot of these like front end node based models to using
- 55:56anything from, you know, from, you know, the models that you
- 56:00just talked about 112.2, you know, there there's so many out
- 56:04there to to figure out what you want to do.
- 56:07And each one of them specializes a little bit differently in
- 56:12terms of like, you know, is it doing more of a style again?
- 56:14Or is it using, you know, just just straight up, you know,
- 56:18diffusion model in in terms of how it's working with its
- 56:21output. And also back to the big thing
- 56:24of like what all this data is trained on and how it's trained
- 56:28and how it continues to be trained.
- 56:31And that's a big thing for everybody because rightfully so,
- 56:34a lot of these platforms have scraped all of media, all of
- 56:39YouTube a lot. All of them.
- 56:41I don't see any other way for them to have been able to
- 56:44generate all of these. Incredible.
- 56:46Videos and we're not, none of us are lawyers here, but there
- 56:50there is that ethical question too.
- 56:53And there's a few models out there and companies that are
- 56:55trying to do better by saying, hey, we have ethical models or
- 56:59we've been trained up to a point and from this point on it's
- 57:02ethical. But you know, and this is back
- 57:07to education is like if you're working with clients and you're
- 57:10working with studios and you're doing things with, you know, IP
- 57:15and copyright holders or trademarks, you have to be aware
- 57:19of what's going on with these different tools.
- 57:21You have to look into their terms of service to say like
- 57:23even with Epic is looking at the terms of service, when you use a
- 57:26tool and it's free, you're the end product, your data, your
- 57:30likeness, all of that. They want that.
- 57:33They want our data because they need to build bigger data sets.
- 57:37They need to get access to what we're generating.
- 57:39All this to either train the model because they're running
- 57:44out of data. So This is why we're having
- 57:46these legality issues. And you know, a lot of people
- 57:48are suing each other back and forth on to the like the fair
- 57:52use Gray area where they're trying to go, well, we can't
- 57:57advance, you know, human, human technology without using
- 58:01everybody's data. And my thing is I'm all for the
- 58:05tech. There's a lot of things I do
- 58:07like a lot about, you know, generative AI, the models,
- 58:11things like that, but I'm an artist.
- 58:13I do think that if your data, your art, your music, whatever
- 58:18it is that you've created is in these sets, you should be
- 58:22compensated for it. Like that is just that's like
- 58:26that is the base ground level talk of that, whether you're
- 58:29trying to do whatever, you know, vantage point of saying, well,
- 58:32we can't advance that. And it was like, well, then you
- 58:34can't advance it If you're taking from other people that is
- 58:38wrong. You should pay those people or
- 58:42come up with a license deal for them.
- 58:43Some people are doing this. Yeah, they should pay the whole
- 58:46world at this point 'cause I feel like they've trained on
- 58:49everyone's data. Yeah.
- 58:50And that and that, that is the weird area we're in of the Gray.
- 58:55The Gray area of, of working on studio project things like that
- 58:59is, you know, if you work with a studio and they give you the
- 59:03rights to their library of IP, their technology, and you're
- 59:08training that data locally within a sandbox meeting.
- 59:10Like it's a that's how people are doing things today and how
- 59:14since we've last talked, like a lot of that has been happening
- 59:17or. So film studios, they run their
- 59:20models locally, like Paramount's runs their own media.
- 59:23There's a lot of companies that have been experimenting with
- 59:26this tech for a very long time. A lot of this stuff is not new.
- 59:30You know, if you look at the papers at SIGGRAPH and you go
- 59:34back 10 years, you look at things from Disney Research out
- 59:36of Zurich, like generative and machine learning, computer
- 59:41vision and a lot of other modalities.
- 59:43Because like virtual production, AI is a blanket term.
- 59:45And it's like, what are we talking about here?
- 59:48Or like AI, you talk about robotics, like a lot of this
- 59:53technology is not actually artificial intelligence.
- 59:56It is not intelligent. It is pattern recognition, so.
- 1:00:01There there's back to this thing of like how people were confused
- 1:00:03of like, well, is, is virtual production this one thing or
- 1:00:06whatever it is? It's like, well, let's talk
- 1:00:08about what your project is, what your technology is and what
- 1:00:11you're trying to get out of it. And it's, it's sometimes it's a
- 1:00:14lot of different modalities brought together and it could be
- 1:00:19diffusion models mixed with, you know, other parts of machine
- 1:00:22learning models and, and, or hardware via computer vision.
- 1:00:27A lot of the stuff that we see in autonomous cars when you take
- 1:00:30a Waymo like that is a hodgepodge of AI at work.
- 1:00:35You know, when you look at so many of the stuff that NVIDIA
- 1:00:39has been talking about, you know, GTCGDC and SIGGRAPH
- 1:00:44NVIDIA, I mean, we, we love them because they power so much of
- 1:00:48all we do with their, with their hardware.
- 1:00:50But from the software side too, of like you look at all these
- 1:00:55different models that are out there with like NVIDIA with,
- 1:00:58with Callan, with their diffusion models with Nerf and
- 1:01:01splats and, and robotics and, and you know, all all the
- 1:01:05computer vision stuff out there. Like there's so much in terms of
- 1:01:13what, what you can do with these different tools for not only
- 1:01:16storytelling, but you know, you guys have talked a lot about
- 1:01:19that too, of like just from an entrepreneurial thing is like,
- 1:01:22you know, making a product and that product could be a course.
- 1:01:26That product could be a camera, physical stuff like you can go
- 1:01:31out and we've seen it too, like back to V cams or virtual
- 1:01:33production. It's like you can go out and go
- 1:01:36to grab an Adreno board, you know, grab some stuff from like
- 1:01:40Adafruit and and some other stuff.
- 1:01:41Make your own joy con controllers as your own V Cam
- 1:01:44system wrapped around a phone, an iPad or an Android device.
- 1:01:48And you've got a pretty robust virtual production tool set that
- 1:01:52you didn't. Maybe, you know, a few 100 bucks
- 1:01:56you can do this. And now that you do have things
- 1:01:59like ChatGPT and Claude and Grok and all this stuff like figuring
- 1:02:03out how to make that Blueprint class to talk to this or to like
- 1:02:07say, hey, I'm not really sure of this.
- 1:02:11You know, with like Google Notebook, feed it APDF of a
- 1:02:17manual on or a book on something that you have access to and go
- 1:02:24have a conversation like a podcast with it.
- 1:02:26It's really cool. And go, hey, tell me how you
- 1:02:30would build this Blueprint class to control OSC, to talk to this
- 1:02:34hardware device to trigger an event within the engine and, and
- 1:02:42you can pick part all these different hardware and software
- 1:02:45things. But it brings together being
- 1:02:47able to have these devices and tools talk to each other and
- 1:02:51then how you either as a team or individually interact with these
- 1:02:55tools. Cause you know, I see a lot of
- 1:02:58things too with the generative being like, well, you're not
- 1:03:01going to need a camera in the future, which I think is also a
- 1:03:03ton of BS. It's like my thing is that one
- 1:03:06of the things I'm really excited is to see capture devices like
- 1:03:10iPhones and even cinema cameras become more like world capture
- 1:03:16devices. Just like we have we saw stuff
- 1:03:19with with world's building devices or world generating Is
- 1:03:27there's something still visceral about going out there with the
- 1:03:30camera and shooting like I still shoot film.
- 1:03:33You know, there's a lot of people that still shoot film it
- 1:03:35whether it's still or it's motion, just because digital
- 1:03:39came along. Yes, it it went down in its
- 1:03:43usage because it's still it's expensive to shoot film and it's
- 1:03:46expensive to process it and it's not the greatest environmentally
- 1:03:49sound thing. So we advanced that and we, you
- 1:03:52know, thank you Kodak for for coming out with the with the,
- 1:03:55with the first, you know, digital camera.
- 1:03:59But, but looking at still going out with a, a camera, whatever
- 1:04:05it is, whether it be, you know, Panavision system, a red system
- 1:04:08all the way down to FX3 or your phone, Like people forget just
- 1:04:12how powerful the stuff is in in our pocket or our backpacks.
- 1:04:16As a, like you have a studio in your backpack.
- 1:04:20That's, that's crazy to think that somebody even five years
- 1:04:27ago to say between an iPhone and a laptop and a little bit of
- 1:04:30know how, what you can put out there.
- 1:04:33And then you have a, you have the biggest megaphone through
- 1:04:38social media ever. And even if you have two
- 1:04:41followers today or no followers, doesn't matter.
- 1:04:44Like if you stay consistent, like you like look at how you
- 1:04:48guys have built the community over, you know, since you, did
- 1:04:51you ever think like when you guys started the coffee podcast
- 1:04:54that you'd be like, you know, hovering around, you know,
- 1:04:57200,000 followers on one particular platform and like
- 1:05:03being consistent with the stuff. So like setting the time aside
- 1:05:06to learn what a tool like Unreal Engine can do for you.
- 1:05:10And you know, just just experimenting with different
- 1:05:14things. And like, you may find out that
- 1:05:17like, hey, you either really don't like this part of creating
- 1:05:20things, but you may run into like 5 people that are really
- 1:05:24good at it. They're faster than you at it
- 1:05:26and that's great. Maybe you want to be more of the
- 1:05:29cinematographer or director or animation director or you want
- 1:05:34to be the composer that's just doing all the scores to to all
- 1:05:38this stuff. I mean, people sleep on scores
- 1:05:42in video games. You know, look at that training
- 1:05:45like the soundtracks just on par with with with how good that
- 1:05:50game is or should say that training too.
- 1:05:53You're up to training. Too.
- 1:05:54Most people haven't even finished one yet, but yeah,
- 1:05:56yeah, shout out, shout out to that amazing team over there,
- 1:06:00Kojima. I think one of the things that
- 1:06:03is really prominent in my head when it comes to the discussion
- 1:06:08of AI and where it's all going, in my opinion, I feel like being
- 1:06:14a generalist is far more important these days than being
- 1:06:18a specialist. Because if you are, let's say,
- 1:06:22only focused on 3D modeling and only 3D modeling and you're not
- 1:06:27willing to go and learn a little bit of texturing, learn a bit of
- 1:06:30animation, some lighting composition.
- 1:06:32And if you got what you want to build with what you know, AI is
- 1:06:36slowly coming in and mastering each and every one of these
- 1:06:39disciplines. So right now, if I compare this
- 1:06:42to 2-3 years ago when I first came to SIGRAF, I remember
- 1:06:44clearly I went to a talk by Tim Miller, guy behind yes, love
- 1:06:50this robots. And he literally talked about
- 1:06:52mid journey and that was the first time I heard about it and
- 1:06:55I was like what what? What the fuck is mid journey?
- 1:06:57And that time was Discord problem, yeah, on this score, no
- 1:07:00interface, there was nothing. Else a couple of years ago,
- 1:07:03yeah, like probably like version 2, version 3.
- 1:07:06Yeah, we had no video generation.
- 1:07:08None of these were there. A lot of people didn't even know
- 1:07:11about image generation. I didn't know about it at the
- 1:07:13time. And we were making meta human
- 1:07:14tutorials. I remember I went back home, I
- 1:07:17checked out my journey and the first time I generated an image,
- 1:07:20it was the most beautiful moment because I could, I couldn't
- 1:07:24believe my eyes that I could write a prompt and then an image
- 1:07:27was generated. Fast forward to today, we have
- 1:07:31image to 3D models that are pretty damn good.
- 1:07:35We've tested them. It's insane.
- 1:07:37And they're now adding quadri topology to it with AI.
- 1:07:41They're not adding PBR textures. It's not.
- 1:07:43Perfect. No, it's not there yet.
- 1:07:45But yeah, you know, you look at tools like Spark 3D and NVIDIA
- 1:07:48part Packer and you know, I've been messing with those heavily
- 1:07:52to just see where we are of, of that like image to or the
- 1:07:57different models of like, you know, you've got image, image,
- 1:08:00image to video, image to 3D. You have a lot of different
- 1:08:03flavors out there. But it's always, you know, it's
- 1:08:06always like the gut check of like, all right, is it good
- 1:08:08enough yet? And like every time you come
- 1:08:10back to it, our new model update happens, something like that.
- 1:08:13You're like, oh wow, like. It got better.
- 1:08:16Maybe a little bit of work, a little bit of retop all getting
- 1:08:19it in the hands of a seasoned artist.
- 1:08:21And even from like character generation, you know, you'd
- 1:08:25probably still hand that off to a rigor or to a seasoned artist
- 1:08:28to be like, or you get the nuances of learning in that.
- 1:08:31And that's the the other side of it too, is like a lot of these
- 1:08:35models don't have like the the exact finite controls that we
- 1:08:39need for tweening and animation. Some of this stuff is getting
- 1:08:43there. We're seeing papers talk about
- 1:08:45that. And yeah, we saw the stuff from,
- 1:08:49you know, our buddies at, well, formerly Wonder Dynamics.
- 1:08:53Studio now. Flow Studio and and, you know,
- 1:08:57move AI and and you know, markerless mocap stuff with
- 1:09:01like, you know, capture and Moven 3 and lots and lots of
- 1:09:06things out there, you know, from and they and they hit the
- 1:09:09sliding scale of of of cost too. Some of this stuff, you know,
- 1:09:13it's still either expensive behind pay wall that you have to
- 1:09:18subscribe to monthly, yearly, whatever it is like, you know,
- 1:09:21getting access to to Vaio 3. You know, it's still pricey.
- 1:09:26You know, some of these these things add up pretty quick and
- 1:09:30you go, well, would it have been faster just with like a a team
- 1:09:33of seasoned artists? And sometimes that is the answer
- 1:09:36because the the pain points of the gaps that are still there in
- 1:09:41in these diffusion models in these certain platforms still
- 1:09:45show that we're still at a very early stage of this stuff.
- 1:09:49And like you said, we are seeing these different things come out
- 1:09:54and you're like, the big thing is pay attention to this.
- 1:09:58Yeah, yeah. No, it's very main thing this
- 1:09:59is. The thing, yes.
- 1:10:00So for me, I understand at this moment in time, it's not there
- 1:10:04yet, but I, I will swear and we will come back to another
- 1:10:08podcast. There will come a time where
- 1:10:11they are at a certain point where any artist can just pick
- 1:10:15them up and use them in a professional work and no one
- 1:10:18will be able to tell the difference if it was done by a
- 1:10:20human artist or an AI artist. I think he will get there.
- 1:10:23Which is why I want to get to my next point and say, although
- 1:10:26this is scary, this is what I would recommend to people
- 1:10:29because this is the approach I will take myself.
- 1:10:33I feel like if you're just going to stay in a 3D modeling lane or
- 1:10:35a 3D texturing lane, this is where it could be a little bit
- 1:10:38scary because those specific parts of the workflow could be
- 1:10:43as at a certain point in time, not replaced by AI, but
- 1:10:46automated. Which means most companies might
- 1:10:48look into that if it's cheaper, faster.
- 1:10:51However, if you're someone like Corey Williams, where you're not
- 1:10:54focusing on a specific part, you're actually coming up with
- 1:10:57ideas. I'm going to make a show.
- 1:10:59It's not about the 3D model, it's not about the 3D texture,
- 1:11:02it's not about the animation, but the vision.
- 1:11:04About the vision entertaining people here.
- 1:11:08No AI can replace Corey because people want to go to YouTube to
- 1:11:11watch Silly Crocodile by by Corey and for me it's going to
- 1:11:15be very hard to replace him because unless AI is going to do
- 1:11:18exactly what he does again, people are watching that for a
- 1:11:22specific reason it's the story that he's telling it's.
- 1:11:25Yeah, every single magic of like, like you said with with
- 1:11:28Tim and the team at Blair and you know, teams at, at, at at
- 1:11:32Sony and Pixar and Frame store and all, you know, and ILM and
- 1:11:36all these amazing people to their like a lot of them have a
- 1:11:41good core understanding of storytelling.
- 1:11:43They have a good core understanding of the the basis
- 1:11:47of what these technologies can and can't do.
- 1:11:50If you happen to be lucky to work at the these studios and
- 1:11:55you know, the big studios have R&D departments, you know, they
- 1:11:59have been working in research and development and and and kind
- 1:12:03of testing these things out on different productions.
- 1:12:05I mean, you look at the, the breakdowns for the multiple
- 1:12:08seasons of, of what they did for Love death and robots with
- 1:12:11animation and Tim and the team, obviously Tim is very vocal of
- 1:12:16their introduction of using Unreal Engine more and more,
- 1:12:20which something that may have been Maya Max or a different
- 1:12:26type of traditional workflow. And it's not to say that, you
- 1:12:29know, there's like, oh, don't use Maya anymore.
- 1:12:32Don't use a particular tool anymore.
- 1:12:34It's it's always going to be in that evolution of that.
- 1:12:38And to your point of having that kind of generalist idea or the
- 1:12:42specialty those were because at that time technology was in that
- 1:12:48place that made us have to rig a certain way or have to, you
- 1:12:53know, go into Zebra, sure have to go into, you know, Meshy or
- 1:12:57any of these other tools that are out there.
- 1:12:59But as technology has advanced and we found solutions for that,
- 1:13:04it's like, well, we found a solution for that.
- 1:13:06Now we're over here. If you look at the history of
- 1:13:09film, that has been the case since day one, and you look at
- 1:13:12the history of visual effects, which visual effects is much
- 1:13:15older. It's right there with the the
- 1:13:18history of cinema. You have, you know, visual
- 1:13:22effects right there, magic Lantern.
- 1:13:23You have very early forms from back in, you know, the 1800s and
- 1:13:28the birth of film of like special, special and visual
- 1:13:31effects as it goes digital. Visual effects obviously came on
- 1:13:35much later with the advancements of computer generators imagery
- 1:13:40which we saw Ed Catmull who was who did his talk and.
- 1:13:44Did you go to the talk? I missed the talk, but we.
- 1:13:46Missed it too unfortunately. Yeah, Andrew posted a photo at
- 1:13:49his talk. Yeah, and Ian from Before and
- 1:13:52Afters always does good follow-ups on that stuff of of
- 1:13:56of Yeah, he was at that talk. I think even he's been hosting a
- 1:13:59bunch of talks at at the festival.
- 1:14:01And yeah, you just look at these these these shifts as that these
- 1:14:10solutions which were problems before someone's going to find
- 1:14:13out hopefully a solve to it. And if you look at the history
- 1:14:18of our tools from editorial, post production, visual facts,
- 1:14:22animation, even just from photo manipulating Photoshop,
- 1:14:28Lightroom, you know, all the open sourcing stuff, it's, it's,
- 1:14:32it's amazing what you can do today.
- 1:14:34It's just like I remember having to the process of even
- 1:14:38colorizing black and white to color and like the amount of,
- 1:14:41you know, channel mixers and, and layers you would have to do
- 1:14:45and sitting there with RGB curves to get things like now
- 1:14:48it's instantaneous, you know, and it's like the solves are
- 1:14:53going to happen and it's, it is moving forward, whether you like
- 1:14:57it or not. And this isn't to knock any of
- 1:15:00the people that are concerned about the their stuff being
- 1:15:04stolen and, and included into these data sets.
- 1:15:06That's you have every right to question that.
- 1:15:09Absolutely. And to be upset about that,
- 1:15:11especially if you have been an artist.
- 1:15:13Likely all of our stuffs already in these.
- 1:15:17No doubt. But, but at the end of the day,
- 1:15:20it is, it is not the guarantee that like that role that was
- 1:15:28there previously will be a role that will be there in 5 to 10
- 1:15:32years because we are just evolving in lots of different
- 1:15:38spaces here. And the hard part to be like,
- 1:15:41well, I'm never going to use whatever it is.
- 1:15:43There's some people they're probably like, hey, I, I don't,
- 1:15:45I don't want to learn an Unreal Engine.
- 1:15:47I'm not going to use that. I'm like, I'm, I'm in this camp
- 1:15:51and you've had that. You have your very loyalists
- 1:15:52that are just like, I'm a Maya guy or gal.
- 1:15:56Well, I'm not going to learn this new tool.
- 1:15:58It's a little bit of muscle memory too, to be like, you are
- 1:16:00so fast and efficient on a certain tool.
- 1:16:03Like the first time I started using Unreal Engine I was like.
- 1:16:08I never. Mind I.
- 1:16:09Remember, it's like how you feel in Houdini today.
- 1:16:11Yeah, it's just like whoa. Or even Touchdesign or Notch or
- 1:16:14any of these stuff. So many buttons, so many things
- 1:16:17to. Click on.
- 1:16:17You can learn that. Like that's the thing.
- 1:16:20You can learn that thing. You can be taught it.
- 1:16:23You can figure it out by a happy accident.
- 1:16:25See of like, you know, how you're mixing nodes and, and
- 1:16:29trying out different things? Like you can figure that out.
- 1:16:32It's the other side of it that is much harder if you're jumping
- 1:16:38into being so used to prompting everything.
- 1:16:42And, and that's the thing, that's the scary part.
- 1:16:44Yeah. If you're going to learn about
- 1:16:45prompting, learn about the physical side of creating too.
- 1:16:49You know, it's just like the way we are right now, just as like a
- 1:16:54human race is there's a lot of things that we see and we suck
- 1:17:00into our brains and that is into our collective.
- 1:17:02So I've seen a lot of that talking to be like, well, humans
- 1:17:04are just like AI or, or generative.
- 1:17:07And it's just like, wait, it's much different.
- 1:17:09And it's, it's very different of like how we take things in,
- 1:17:15recompile them and put them out there.
- 1:17:17And yeah, there are always pieces of that from like
- 1:17:20nostalgia or just from what we like, you know, so you may hear
- 1:17:24a music artist that you really like.
- 1:17:28We're all influenced by what collectively we've been through,
- 1:17:32what our surroundings are, what our friends and family are into
- 1:17:36all of that, what trips you may take.
- 1:17:38And you get sparks of ideas of things.
- 1:17:41That's the cool part of it, you know, it's like I, I'm less
- 1:17:48afraid of these technology shifts.
- 1:17:50And it's probably because I've gone through so many of my
- 1:17:52players so far, so maybe. You experimented.
- 1:17:56You pivoted a lot. I think that's why that you're
- 1:17:59so forward thinking and not so scared of at this stage because
- 1:18:04you've been through different stages but with lower speed.
- 1:18:07Now it's just the same thing with the higher speed.
- 1:18:09Yeah. And I and, and I think that as
- 1:18:12people progress over whatever career they take, they're going
- 1:18:17to go through these ups and downs of, of, you know, what the
- 1:18:21pain points are, where the, the solve comes in.
- 1:18:25And you know, there's not always going to be the quick answer.
- 1:18:29As fast as AI and generative and machine learning computer vision
- 1:18:34are scaling up for us, I think too what we're going to see is a
- 1:18:38little bit of a reset that that bubble will pop a little bit
- 1:18:42because we've, we've all seen these hype cycles and what I,
- 1:18:46what I still feel is going to happen because I think it's true
- 1:18:50to today, we all go see our favorite actors or our favorite
- 1:18:54musicians and we seek out other humans for connection.
- 1:18:59So even though you can have a digital twin of Ryan Gosling,
- 1:19:04you're probably going to be less likely to be like, am I Internet
- 1:19:08or do I want to see the real, you know, yeah, like you want to
- 1:19:12see the funny quips and the quick things or the Ryan
- 1:19:14Reynolds and you get those quick reactionary Deadpool like,
- 1:19:19because it's that it's that human thing that we're we're
- 1:19:22searching for that that human connection.
- 1:19:23And you know, I even think music is even 100.
- 1:19:27Percent. Way crazier than that he was.
- 1:19:29Telling me, actually he was telling me that.
- 1:19:31So we went and saw a few DJ our favorite D's and he was telling
- 1:19:35me if you tell me tomorrow there is an AI model of for example
- 1:19:38Animal that plays the exact same music with the same style.
- 1:19:42I still prefer if that exists today.
- 1:19:44Right now, which it probably does, I mean like you could go
- 1:19:46to to you know, you could go to 11 labs, you can go to you can
- 1:19:50go you could feed these things or you can train your own local
- 1:19:54control net or or whatever it is in terms of models and get
- 1:19:58something to sound very similar to like that or almost like an
- 1:20:01exact replica. But it's like, you know, if
- 1:20:03you're going to go see. Dead Mouse, you're going to see
- 1:20:07the show, you know, like you're going to see a live event.
- 1:20:11You're going to see. Humans.
- 1:20:13There is something about the human element that we seek.
- 1:20:16Yeah, and there's another part too, about just like our brains
- 1:20:20in cinema being trained for over 100 years of like knowing that
- 1:20:2524 and 20 fives per second is that is filmic to us.
- 1:20:28Like it it is a is a human thing, persistence of vision.
- 1:20:32We know things that are like 30 and 60 frames are more, you
- 1:20:36know, they're, they're more in a, you know, real life.
- 1:20:40Live stream style. Yeah, like video.
- 1:20:42Games would be like 60 and above.
- 1:20:43Yeah, it's, yeah, it's, it's, it's like a visceral texture
- 1:20:46thing too, of like just what our brains have been trained on, you
- 1:20:50know, for over 100 years. And it's the same thing with,
- 1:20:53you know, with, with video games.
- 1:20:55It's like you can crank down video games to 24, but that's a
- 1:20:59lot slower than what people are. Used to.
- 1:21:01Would it be fun? No, not reactionary time dude.
- 1:21:04It'd be like man that that snipe shot is taking really long for
- 1:21:08those costs playing. Competitive games like that.
- 1:21:11Yeah, yeah, yeah, yeah. You want.
- 1:21:13Yeah. If you could feed everybody at
- 1:21:14240 frames per second, they would take it all day long.
- 1:21:17Yes, but they would just maybe want a a a style GAN over that
- 1:21:21that would like simulate the film look on top of the real
- 1:21:25time playback which they have. So we have you know, we have,
- 1:21:29you know, stylistic shaders. Our buddy Michael Neely over at
- 1:21:32Epic who did a whole talk and it has this whole thing up on on
- 1:21:38the Epic games YouTube of like literally just specializing help
- 1:21:42us for stylistic shaders because they want that kind of tune
- 1:21:46shaded stylistic comic effect. And you're still going to have
- 1:21:50people that really know how to write shaders and that's going
- 1:21:54to get faster. So, you know, you'd have that
- 1:21:56like, and I think spider man, you know, in terms of like
- 1:21:59really art directed looks that the team had painstakingly used
- 1:22:09and you know, both in real time tools and post rendered tools to
- 1:22:13get that stylistic look. But like really those different
- 1:22:16spider man's that we had, you know, in the the, the last
- 1:22:21spider man, not the Tom Holland Spider man.
- 1:22:24The spider verse, right. The spider verse.
- 1:22:26Yeah, as you look at the direction of that, there's so
- 1:22:30many different styles in in that for the for the different spider
- 1:22:35verse in there, 'cause it's like, well, everybody has their
- 1:22:38own story based off of which spider verse they're in and
- 1:22:41their style and art direction for all that made made the book
- 1:22:45awesome. It made the film awesome.
- 1:22:47And you you get really good music with that.
- 1:22:51It's like, and that's, that's the specialty of, you know, that
- 1:22:55kind of collaborative film makings.
- 1:22:56You're bringing all these people from different backgrounds that
- 1:22:58can have input into that which make this special thing of the
- 1:23:01output. If if it had just been
- 1:23:04generated, it would look very different.
- 1:23:06It's story wise it'd be very different and I'm sure the music
- 1:23:09would be different. That's very true.
- 1:23:11And it goes back again to the vision that we discussed, the
- 1:23:14vision of the director and the vision of the team behind it.
- 1:23:17Jim, I want to ask you, if you were to start your career today
- 1:23:22knowing everything that you know now, what would you do?
- 1:23:26I mean, I would dive into this stuff.
- 1:23:28I don't I don't think that me kind of always being like driven
- 1:23:33towards art and music and wanting to create no matter what
- 1:23:36it was is like I've always just circled around to that.
- 1:23:39So if I started today, as daunting as things look in terms
- 1:23:44of like there's so much out there between where do you go?
- 1:23:47Do you go to Unreal Engine? Do you go to Houdini?
- 1:23:49Do you go to Blender? Do you go to Maya?
- 1:23:51Do do you go all heavy? Just generative?
- 1:23:53Because there's some people that are starting today and they're
- 1:23:56learning that the basis is like, it's Blender, it's Unreal
- 1:24:01Engine, it's ChatGPT, it's, you know, any of these generative
- 1:24:04platforms out there is like, First off, you're going to start
- 1:24:08with like learn about storytelling, like without that
- 1:24:12core base of, of like story structure or shot composition or
- 1:24:17you know how to do, I'm not saying like learn the box to
- 1:24:20stay in the box. You're supposed to break
- 1:24:22out-of-the-box and try and different things break the
- 1:24:24rules, but like you need still need to engage people into this.
- 1:24:30So like the to the technology is a formality, It's a means to an
- 1:24:33end. You know, for all of this stuff,
- 1:24:35it doesn't matter if you like, you know, you never touched
- 1:24:39Unreal Engine or you never touched, you know, ARA 235, you
- 1:24:45know, film camera, like going out there and creating, going
- 1:24:49out there and diving into these different tools because you guys
- 1:24:53get it too, is like asking where, where do we start?
- 1:24:56And it's like, well, if you're going to pick one thing, pick
- 1:24:59Unreal Engine, you know, pick blender, you know, see what
- 1:25:04these tools can do because they're all progressing.
- 1:25:08And the start off small, like, you know, take a course, watch a
- 1:25:13YouTube video, look on Instagram, check out hashtags,
- 1:25:16whatever it is, you'll find something to be like, like like,
- 1:25:20hey, I like this. Go recreate that thing that you
- 1:25:23like. Learn how to do it 5 different
- 1:25:26ways. You know, learn about just
- 1:25:29placing an object, a camera and a light one, one and one mess
- 1:25:34around with it, Try out different lenses.
- 1:25:37Look at what you know, you know, a spot versus rec light does,
- 1:25:41you know, look at how the subject lights.
- 1:25:44And then, you know, go back to, you know, painters and artists
- 1:25:49and look at like how much those guys and gals studied light and
- 1:25:53and. Shadows.
- 1:25:54Yeah, light and shadows. I mean, that's the big sell.
- 1:25:56Like we, we've all seen bad, you know, things that are shot on
- 1:26:00LED that look like, you know, they're the subjects, you know,
- 1:26:05are popping off of, you know, their background.
- 1:26:08A lot of that has to do with lighting and color blending.
- 1:26:12So like having all these bases of like getting into things
- 1:26:16today is like, you know, information is out there.
- 1:26:20It is so readily available. When I started, you had to read
- 1:26:23books. There was there was no YouTube.
- 1:26:26I cannot imagine how that must. Have felt like imagine manual of
- 1:26:29blender. I had books on books of like
- 1:26:33learning, you know, Avid and learning Final Cut and learning
- 1:26:38and learning combustion. And you know, early days of like
- 1:26:41when when Nuke came out and I was just like, yes, because it
- 1:26:45was like, you know, I went from, you know, combustion to shake to
- 1:26:48Nuke. And then obviously we had After
- 1:26:51Effects and we had Photoshop and we had early days of premiere
- 1:26:53and premiere was on Mac. Then it wasn't on Mac.
- 1:26:55Then it came back. And like, and Davinci Resolve
- 1:26:59used to be a system that was hundreds of thousands of
- 1:27:01dollars. You know, flame and smoke
- 1:27:03systems were millions of dollars, especially if you were
- 1:27:06outfitting a whole thing with storage.
- 1:27:08And you know, it's all this proprietary stuff now.
- 1:27:11Like we say, we've got, you know, edit and cap cut and and,
- 1:27:16and, and, you know, black magic tools and things like that that
- 1:27:19run on our phones and our iPads and you know, like you're the
- 1:27:24studio that we're in today. Just 10 years ago, you know,
- 1:27:28studio like this, even for a podcast and all that it, it
- 1:27:31would have cost 10 thousands, if not, you know, millions of
- 1:27:34dollars to outfit everything like that.
- 1:27:36So, you know, the, the biggest thing I can say to people is
- 1:27:40just like, like education is key.
- 1:27:42Like you should always be in learning mode, Always like know
- 1:27:47when to like keep your ears open and your mouth shut in terms of
- 1:27:51like just be a sponge of times. And then knowing when to like
- 1:27:54collaborate and kind of speak up to be like, oh, I have these
- 1:27:57ideas and just like kind of throw stuff at the, at the
- 1:27:59dartboard. But yeah, it's, it's, it's a
- 1:28:05awesome time for people that are like starting out now or
- 1:28:08pivoting because it's like, what do you want to do?
- 1:28:12There's there's some way to beg, borrow, steal, or buy?
- 1:28:17I love that. These things out there to kind
- 1:28:20of like do make a product, make a course.
- 1:28:25If you want to develop a camera, you can develop a camera.
- 1:28:27You can learn about Fpgas and ASICS and you can you can go buy
- 1:28:31adrenal boards and and hook up machine learning, You know, a
- 1:28:35databases to whatever it is. Like it's really the Sky's the
- 1:28:40limit into what this what it is that you want to do.
- 1:28:45And I know it's daunting and people are kind of like scared
- 1:28:48about we're in this big shift. Like obviously there's a lot of
- 1:28:51things that have been going on in the industry in terms of like
- 1:28:55feature films and episodic and even AAA games.
- 1:28:58Like we've all seen these studios closed down, a lot of
- 1:29:02shrinking budgets have shrinked. But these are all things to just
- 1:29:07go, hey, you got to pivot and like if if you don't work at a
- 1:29:12studio anymore, make a studio, even if.
- 1:29:16It's just you. Absolutely no, that's so true.
- 1:29:18And The thing is, I watched Gareth Edwards talk about this.
- 1:29:23I saw Christopher Nolan talk about this.
- 1:29:24Did you say just go and fucking shoot something?
- 1:29:26Just grab a camera. Ridley Scott and his brother
- 1:29:30start off making short films. Christopher Nolan, Gareth
- 1:29:33Edwards and I remember early days with Gareth on like FX
- 1:29:37Guide and FXPHD. There's some other people that
- 1:29:40were on like some of the old communities back in the day
- 1:29:43where you're just like, look at, look at where they went.
- 1:29:47Obviously you got to have the drive, you got to have the
- 1:29:49passion. Like people who succeed in kind
- 1:29:53of any industry, something is driving them.
- 1:29:57They have that vision and there's going to be ups and
- 1:30:01downs. Like I've had tons, like I've
- 1:30:03had projects that we finished never seen the light of day.
- 1:30:08And it's painful and it's painful for you have all of
- 1:30:11these people that spend time away from their their family,
- 1:30:14their lives, all this to like, you know, come together to make
- 1:30:16films come make. We hear this all the time
- 1:30:19recently with game studios shutting down and people put
- 1:30:22games take even longer than. Films they put 2-3 years into
- 1:30:26the project and if they never see the light of day, that's 2-3
- 1:30:28years of your life gone. You got paid but then.
- 1:30:31No, but you care about the work people see, the fulfillment of
- 1:30:33people seeing your work. Yeah.
- 1:30:35And the enjoyment to be like, not only that, but, you know,
- 1:30:38like it's fun that I still get to show, you know, my family, my
- 1:30:42kids, my wife, like, you know, take them to the theater to see.
- 1:30:47This is what I worked. On it's like, hey, if dad is
- 1:30:49gone for three months, like, yeah, Dad, this better be a good
- 1:30:54film spent this much time away. But you know, and there's there
- 1:30:59is the sacrifice. It's like, you know, like there
- 1:31:02you are giving up stuff, whether you're an individual or, you
- 1:31:06know, your family or that there's always this give and
- 1:31:08take too. And it's at, you know, you're
- 1:31:10sacrificing things for that passion to be like, like, I, I
- 1:31:15want to be a filmmaker. I want to be, you know, I want
- 1:31:18to make my own game. Like they, we go back to this
- 1:31:21thing too of like today, there's no excuses.
- 1:31:24Like there's literally, you brought it up earlier of the
- 1:31:27like, hey, I don't have access to this.
- 1:31:29Like somebody may not have a laptop, they may not have an
- 1:31:33iPhone that can run, you know, any of the, the tools that we're
- 1:31:38talking about of interfacing with like Meta human animator
- 1:31:40and all this stuff. But like there's a means.
- 1:31:43Do you know someone that has an iPhone or a a gaming laptop or
- 1:31:46something like that or? You can hustle your way to get
- 1:31:49one. I mean, that's what I did as
- 1:31:50well. I wanted some monitors, extra
- 1:31:52monitors for my streaming. I sold my assignments and essays
- 1:31:55to my other students like and they wanted you.
- 1:31:58Figured it. Out, yeah.
- 1:31:59You you can go what you figure it out you.
- 1:32:01Can go work at a coffee shop for three months and save enough
- 1:32:04money to buy a second hand phone that can run whatever we're
- 1:32:07talking about. There are ways just to get your
- 1:32:10foot in the door and once you have the hardware, Unreal Engine
- 1:32:14is free, Blender is free, and you can pretty much do anything
- 1:32:17that all the big studios do alone with Unreal Engine and
- 1:32:21with Blender. It's just going to be obviously
- 1:32:24there. There is a, there's a different
- 1:32:26in terms of like outputs and all and all that, but it like we've
- 1:32:29talked about a few guys and gals who are putting out amazing work
- 1:32:33as individuals. Like they don't have a studio.
- 1:32:36They don't even have, you know, anyone else helping them.
- 1:32:39They're teaching themselves this.
- 1:32:41They're failing because, you know, you talk to half, half our
- 1:32:44friends that are doing stuff in, in, in these different tools and
- 1:32:47they're just like asking the same questions.
- 1:32:49We got caught up like, wait, how do you do this in blueprints or
- 1:32:54how do you do this in, you know, in in chaos or Niagara or
- 1:32:58whatever it is? And you, you have to be
- 1:33:01persistent at all of this stuff. And if you have a vision to like
- 1:33:06make a film or make a cinematic or all you do is like short
- 1:33:09experiments. Like I say, if you want to make
- 1:33:13a film, great. But in the meantime, just do
- 1:33:16little shorts, like just be really quick, like learn, learn
- 1:33:20how to get an in and out of a tool really quick And do one
- 1:33:23thing like at that at the end of a week or at the end of the day,
- 1:33:28like give yourself a goal to just go, I'm going to learn how
- 1:33:32to write an editor utility widget that makes my outputs
- 1:33:36through MRQ faster or, you know, or, or beat or graph, you know,
- 1:33:40it's like, or learn that thing like I'm going to learn how to
- 1:33:44do a doughnut in Blender today. I actually love that because if
- 1:33:48you give yourself a deadline and a goal, it's very easy to follow
- 1:33:53when you make it open, when you don't have a deadline or you
- 1:33:56just say, oh, I want to learn, Andre, but what do you want to
- 1:33:58learn? There are so many things you.
- 1:34:00Learn. That's where it's daunting.
- 1:34:01Yeah. That's where it is hard.
- 1:34:02Yeah. And you don't complete.
- 1:34:03You never start. You need to give yourself a
- 1:34:05specific task at a specific time and make sure the timeline is
- 1:34:09tight because if you give yourself a lot of time, then you
- 1:34:11would say, oh, I can do it tomorrow, I can do it the day
- 1:34:13after. Give yourself a tight.
- 1:34:15Yeah. No, a specific task and do it.
- 1:34:18And then you can do so many mini project and at the end of the
- 1:34:21year you will see I've done 6 mini projects, 7 mini projects.
- 1:34:24Now I can combine all of these skills together and make
- 1:34:27something big. And now you're experienced.
- 1:34:29And one thing, if you have the vision for the film that you
- 1:34:32want to shoot, the short, the game, the commercial, whatever,
- 1:34:35go and do that. If you don't have a you did
- 1:34:37mention this. A remake is always the best
- 1:34:40answer. You just whatever you love,
- 1:34:42whichever director you love, whichever producer, whichever
- 1:34:45game dev follow their lead, remake whatever they made, maybe
- 1:34:49in a smaller portion. So it's not too much for an
- 1:34:51individual because when you remake something, first of all,
- 1:34:54you don't have to think of a vision.
- 1:34:56You're following in someones footstep.
- 1:34:57And more often than not, it's a master that you're following,
- 1:35:00which means you'll train your eye to follow the same
- 1:35:03compositions, the same lighting techniques.
- 1:35:05So when you do end up with a vision now, you'll be hopefully
- 1:35:10as good as them or maybe at least close to them, which means
- 1:35:12your end product is going to look really.
- 1:35:14Good. And you and you also help peel
- 1:35:16back the understanding in your research of like, oh, how did
- 1:35:19they do that? Like, you know.
- 1:35:20Reverse engineer. Yeah, and that's a big thing is
- 1:35:22like I, I do it for fellow cinematographers is like, I'll
- 1:35:27look at a Greg Frazier Dune frame and I'll go, how quick can
- 1:35:33I build that in Unreal Engine? I'll take that one frame or if
- 1:35:38it's a sequence to look at, like testing out new animation tools
- 1:35:41or whatever it is. And I'll go and I'll, I'll start
- 1:35:44backwards into that and I'll be like, well, OK, I know Greg did
- 1:35:48a podcast on this, you know, with Epic or he did one with
- 1:35:52American Cinematographer. And then I, there's his lighting
- 1:35:54plot. Oh, well, I've, I've got his
- 1:35:56lighting plot from him and his gaffer.
- 1:35:58I know where his point sources are.
- 1:36:01Did he use natural light? Did he fill it with a bunch of
- 1:36:03like, you know, vortex aids or did he, you know, did you use a
- 1:36:06wall of 50 vortex eights? Like can I simulate that in the
- 1:36:10engine? And it's like, well, I know what
- 1:36:11the lux values are of that particular light.
- 1:36:14I know how much intensity that's going to be.
- 1:36:17Got to figure out the math on the Unreal Engine side because
- 1:36:21our our our good friends at Epic are working on it.
- 1:36:24But photometrics of 1 to one are not in a lot of these tools yet.
- 1:36:28So you have to figure out like where the sweet spot is for some
- 1:36:32of these tools of simulating lighting that is built one way
- 1:36:38for physical production and different for virtual production
- 1:36:42or for just the way a particular renderer will work.
- 1:36:47So, you know, obviously if you're working with a post
- 1:36:49deferred renderer versus real time rendering, we talk about
- 1:36:52this with real time ray tracing, with patch tracing and and lumen
- 1:36:56and and Unreal Engine and you go that does a funny thing like
- 1:37:02lights in the real world don't do that.
- 1:37:03So you so you kind of understand where you can push and pull
- 1:37:07things in there. And it's really good to use this
- 1:37:11example of taking a, the famous painting a, a film, a commercial
- 1:37:19and then rebuilding that because now you're understanding like,
- 1:37:22Hey, you like that for a reason. Yes.
- 1:37:24Like it's in some way that thing is speaking to you.
- 1:37:28And to recreate that is a good way to understand it.
- 1:37:32But in my my long time good buddy Andrew Kramer.
- 1:37:35Oh yeah, brought this up way, way long ago of like I talked to
- 1:37:40him. I was like, man, I see your
- 1:37:42stuff everywhere. They RIP it off.
- 1:37:43I saw it all over over Vegas and he's like, dude, I don't care.
- 1:37:46He's like if his goal of that stuff and there's been a lot of
- 1:37:50friends that have done, you know, training in that way.
- 1:37:53We talked about like my other buddy Ron Rabinowitz and and
- 1:37:56Winbush and like all like you guys and everybody and taking
- 1:38:01that example, there's nothing wrong with it.
- 1:38:03It's like it's done all the time.
- 1:38:05But like then what you're supposed to do with that is take
- 1:38:08those learnings and then make it your own.
- 1:38:10Absolutely. That's the big thing is like I,
- 1:38:15I don't want to be Greg Frazier, you know, and that's no doctor
- 1:38:19Greg. He's an amazing person and
- 1:38:21amazing talent, but I want to take what I what I like that he
- 1:38:26does and, and, and the, and learn from that and then and mix
- 1:38:30him with that. And it's like, then I should be
- 1:38:33elevating that. Like I should be like a
- 1:38:35challenge because it's like the same thing for, you know, Roger
- 1:38:39Deakins, you know, it's for, for Greg Frazier, it's for Rachel
- 1:38:43Morrison's, it's your, you know, pick your flavor, You know,
- 1:38:48it's, it's, it's whoever your favorite director or artist or
- 1:38:51musician or whatever it is or your favorite TikTok.
- 1:38:54Yeah, like whoever. Yeah, even the.
- 1:38:56Content, even even the content if from our program.
- 1:38:59A lot of people recreated the cinematic that we made and the
- 1:39:03title sequence, and some of the students went above and beyond,
- 1:39:07created something that we didn't even think about.
- 1:39:10The camera work. Yeah, the camera work is crazy.
- 1:39:13They followed the step. They are creative on their own
- 1:39:16and they they took it to another level.
- 1:39:18And it's great to see it and they love to see it because now
- 1:39:21we know that not only they learned, they took it to another
- 1:39:24level. And I think they would be
- 1:39:25successful. Biggest.
- 1:39:26Compliment to you guys is like like in the hope is you should
- 1:39:32want the people taking your course or anything like that to
- 1:39:36surpass that because it's like, oh, they really got what we put
- 1:39:39out there to the concept and you want you should.
- 1:39:44If you're a good person, you want to see those people succeed
- 1:39:46and do even better. Because it's like then the bar
- 1:39:49gets raised and then you get inspired and then you're like,
- 1:39:52OK, well. Let me I'm competitive.
- 1:39:54About that and being a little and there's nothing wrong with a
- 1:39:56little bit of competitive between our friends.
- 1:39:58No, I think that's where progress comes in.
- 1:40:01That's why you like you are applauding for each other to.
- 1:40:03To make better work, I will say like you make something better
- 1:40:06than me or I'm going to make something better.
- 1:40:07Both of the work is elevated now and it's it feels amazing for
- 1:40:11everybody. I'm curious, now that we've been
- 1:40:13talking about Unreal Engine, of course we know when it comes to
- 1:40:16LED volumes, people use Unreal Engines.
- 1:40:18How else are you using Unreal Engine in the industry?
- 1:40:21Let's say previous and other things?
- 1:40:23That was probably one of the first things that I saw.
- 1:40:27I really like that breadcrumb, you know, out there from from
- 1:40:31using Unreal Engine was getting ideas out really quick in a real
- 1:40:36time. Engine is just you can try
- 1:40:39things. I don't like the light there.
- 1:40:41I don't like the camera. I want to put 30 cameras in
- 1:40:43there. What do I do?
- 1:40:44I want to build a camera array virtually.
- 1:40:48It's great to audition things out.
- 1:40:50You want to try out different lenses.
- 1:40:52You want to try out different shaders, different looks
- 1:40:54different, like a noise pattern or grain.
- 1:40:56You want to simulate, simulate film.
- 1:40:58You want to simulate how something would be on a motion
- 1:41:00base with a car or a plane buck or, you know, you want to bring
- 1:41:05in things for for tech fizz by, you know, using Polycam scanner
- 1:41:11verse or reality capture tools and bring that into engine via,
- 1:41:15you know, OBJ, you know, FBXUSDGLTF, like you've got so
- 1:41:21much of it there. And I think that's too people
- 1:41:24that have gotten into Unreal Engine and real time tools are
- 1:41:28just seeing like how quick you can be with things and how quick
- 1:41:32you can collaborate. Because it's like if you go
- 1:41:34into, you know, multi user session and somebody's in like a
- 1:41:38vision pro or a meta quest, If you really want to envelope
- 1:41:41somebody into the world, you can put them still in a virtual
- 1:41:44reality HMD and let them step into it.
- 1:41:47And you do have in real Unreal Engine, you have the updated
- 1:41:51mixed reality tools for collaboration of like world
- 1:41:54building, you know, so if you have a virtual art department,
- 1:41:56you can do it over Zoom, you can do it in, you know, in a virtual
- 1:42:01reality environment. And you can have people, you
- 1:42:04know, operating camera while somebody's moving, you know,
- 1:42:07city scenes around city blocks and different assets in there.
- 1:42:12So really trying out ideas. The cool part about it is you
- 1:42:17may never use Unreal Engine past that point and you may go out
- 1:42:22and shoot on film practically and you may never touch a green
- 1:42:26screen, a blue screen, a Gray screen, or an LED volume.
- 1:42:30But you can do all your tech viz.
- 1:42:34Your pre viz, any of the viz. All the vizes.
- 1:42:38All the vizes. You can do your stunt viz and
- 1:42:42the. Stunt Wiz What?
- 1:42:43Stunt viz. What is?
- 1:42:45That the stuntman. Stuntman.
- 1:42:46Yeah, as if you we use a lot for stunt viz and you know,
- 1:42:51simulating like anything that would be like what we see in
- 1:42:54Mission Impossible or any stunt and John Wick or anything like
- 1:42:57that. And you have a lot of stunt
- 1:43:00coordinators learning on real engine because they may have
- 1:43:04previously had somebody else, a 3D artist that has been doing
- 1:43:07stuff in Maya or Blender or, you know, some other viz program.
- 1:43:12And you know, they're just picking it up because on the
- 1:43:16side like when the stunt coordinators and the stunt
- 1:43:19departments go through, it is, you know, they want to do these
- 1:43:22things because the first and foremost thing is safety.
- 1:43:24So, you know, to visualize this and get buy in.
- 1:43:28This is what's really great about things like Unreal Engine
- 1:43:32is getting buying in early. You can save time and money and
- 1:43:38and safety of people getting hurt to look at that.
- 1:43:42So you're looking at camera angles, you're looking at, you
- 1:43:45know, from a safety perspective, like if you go do a site scan
- 1:43:49with lighter of the set and you know that this wall is here,
- 1:43:52well, you better not put the person next to the wall if the
- 1:43:54car is about to smash into it. So you can look at a lot of this
- 1:43:59stuff early on with, with all the business as we talked about.
- 1:44:04And you know, at the end of the day, Unreal Engine's still free.
- 1:44:08You know, it's like, yes, we have the licensing and we have,
- 1:44:11if you hit the tier level, then you're doing, you know, you're
- 1:44:14doing your Rev share and you pay for them rightfully.
- 1:44:17So that's. A lot.
- 1:44:18If you are making a feature film, you better pay for it
- 1:44:20because you're using it right. Yeah, well, because we want
- 1:44:22those tools to live on. You know, it's like, you know,
- 1:44:26people have to get paid for all the hard work.
- 1:44:28There's a lot of amazing engineers and teams, you know,
- 1:44:31at these different companies that are making these tools for
- 1:44:34us. And I'm more than happy to be
- 1:44:39like, hey, if this tool solves this problem for me and it lets
- 1:44:42me get in and out really quick, hell yeah, I'm paying for it
- 1:44:46within reason. You know, it's like, that's
- 1:44:51what's really cool about these tools and specifically about,
- 1:44:55you know, working closely with the Epic team over the years
- 1:44:57and, and, you know, signing on as an Epic partner.
- 1:45:01I'm, you know, it's like my company, Space play Labs is a
- 1:45:04service partner with, with Epic. I've talked at a bunch of on
- 1:45:08real fast. And I do it because I get
- 1:45:12excited about what freedoms it gives me as a, as you know, as a
- 1:45:18filmmaker, as a director of photography and, and a visual
- 1:45:21effects supervisor. You know, a lot of my day-to-day
- 1:45:25job is solving problems. Like it's either solving
- 1:45:27problems for me or the other directories and DPS and, and
- 1:45:30vendors that I work with. And at the end of the day,
- 1:45:34there's still a lot of people that don't know these tools
- 1:45:36exist out there. So there's still a lot of
- 1:45:38discovery to just be like. There's a lot of education to be
- 1:45:41done. That's the biggest thing is,
- 1:45:43like a lot of, you know, my day-to-day too, is educating
- 1:45:47people on what's out there. You know, it's like, hey, this
- 1:45:51is coming down the pipe, be ready for it.
- 1:45:52But then here's what what's practical and what's here today.
- 1:45:56And, you know, looking at tools like Unreal Engine is really
- 1:46:02just looking at an ecosystem of, and I'm, I'm sure, you know, Tim
- 1:46:09and Kim and Bill and everybody, the big exact set at your world
- 1:46:14building. Yeah.
- 1:46:14Look at look at UEFN, look at Fortnite, look at Rocket League,
- 1:46:18look at everybody building different tools on that.
- 1:46:21And, and look, it's, you know, whatever your flavour is, like
- 1:46:24you can pick other engines out there.
- 1:46:26Just we happen to be a little bit biased towards our engine.
- 1:46:30It it it, it does a lot of what we want it to do.
- 1:46:32No, the updates are so many. Problems, I think one of the
- 1:46:36best reasons, one of the most important reasons why the
- 1:46:40updates are so frequent and also so accurate to what people need
- 1:46:45is because they're trying everything with Fortnite and if
- 1:46:48they can make Fortnite better, all of those updates are sort of
- 1:46:53coming out to us, the artists who may not be working on
- 1:46:56Fortnite, but we get all of the juicy updates.
- 1:46:59For instance, one of the things that I'm super pumped about is a
- 1:47:01new nanite foliage, which means we get to have trees without
- 1:47:06having plains representing the leaves.
- 1:47:08It's actually. Early early days of doing
- 1:47:11foliage of having to, you know, create and speed tree and model
- 1:47:15things in Maya and then bring them in and then optimize them
- 1:47:18to make them perform it. Foliage and weather systems have
- 1:47:21always been a work in progress and a heavy, a heavy load on
- 1:47:27your scenes within real time tools.
- 1:47:29So you're all, you're always trying to figure out, well,
- 1:47:31like, well, maybe that stuff will be cards in the back and,
- 1:47:35or we'll bake that out and then we'll keep this.
- 1:47:36But you know, I had a conversation recently about this
- 1:47:39about like, well, when do you choose to bake something versus
- 1:47:42leave it dynamic? You know, like, why would you
- 1:47:44want something, you know, dynamic versus static or
- 1:47:47anything like that? And when you're working with
- 1:47:49other creatives, especially in, in, you know, whether it's games
- 1:47:53or whether it's feature films or episodic or at at any level is
- 1:47:57you, you know, telling your director, no, they don't want to
- 1:48:00hear that. They want, they want to go like,
- 1:48:02well, how are you solving this problem for me?
- 1:48:04Like if you're suggesting these tools, like, you know, they,
- 1:48:07they get frustrated to be like, wow, I just got this creative
- 1:48:10vision, I want to get it out. And is the tool the limiting
- 1:48:13factor? Like what's the limiting factor
- 1:48:15here? And sometimes it is the tools
- 1:48:17because that's where we are at technology today.
- 1:48:20So, you know, looking just five years ago of what you could do
- 1:48:26in terms of having a dynamically lit scene that you can change
- 1:48:30weather systems and time of day and, and trigger animations and
- 1:48:34things like that. It was really hard, like when
- 1:48:37when we did Pinocchio, you know, years ago.
- 1:48:40It's it's crazy to say years ago now.
- 1:48:43But like a lot of the stuff that that I was working on with the
- 1:48:48teams that I had and with Kevin Bailey and Bob Zemeckis and, and
- 1:48:53you know, the teams that we had there, like it was a heavy lift
- 1:48:57from everybody. And we were breaking new ground
- 1:48:59of like trying out new things that worked and didn't work.
- 1:49:03But we learned the whole time, you know, of, of trying new
- 1:49:07things out and see if it worked. And at that time, we were the
- 1:49:11first feature film to step into, you know, going from like 425 to
- 1:49:16426 and then to 427, especially because we moved into the new
- 1:49:21engine version and working closely with Epic because we're
- 1:49:24doing a lot of image based lighting of like having big LED,
- 1:49:29not volumes, but LE DS on trusts and, and, and set up there
- 1:49:34because of the interactive lighting aspect of it, which at
- 1:49:38that time image based lighting isn't new, but the approach of
- 1:49:43being able to, you know, run DMX and Arnet into real time of
- 1:49:47getting lighting. Even though we had gigantic sets
- 1:49:51and blue screens at that time. You know, there's, there's just
- 1:49:55different aspects of it, but that's all virtual production.
- 1:49:57Even though we didn't have an LED volume, Mandalorian esque
- 1:50:00right behind there for for backgrounds, you know, you can
- 1:50:05use the true LED volume from in camera visual facts of like what
- 1:50:10you're shooting through camera to get for the backgrounds,
- 1:50:12foregrounds and all that. Versus, you know, image based
- 1:50:14lighting of using, you know, pixel mapped through LED or
- 1:50:18through other fixtures, whether it be, you know, cream source or
- 1:50:21whether, you know, sky panels or, you know, pick your flavor
- 1:50:24of cinema lights that are out there that can be addressable
- 1:50:27from pixel mapping. And you have a lot of different
- 1:50:31tools that are all being driven by the engine and the fact that
- 1:50:35people are like, Oh, I didn't even know you could do that with
- 1:50:37you. Can you know, when we when you
- 1:50:40guys show the stuff from Tomorrowland, it's like all that
- 1:50:43stuff talking about all the stuff that like, you know, Joel
- 1:50:45dead mouse does through his shows and and building package
- 1:50:49things for live experiences or building real time interaction.
- 1:50:55And that's the cool thing about it.
- 1:50:56It's like, you know, ADP, your director, art director or even
- 1:51:01even an actor is just like, hey, that thing on the volume or that
- 1:51:05thing on screen. It's it's like, I don't think
- 1:51:08it's right. If it's a fear, all right, get
- 1:51:09rid of it easy. Like we just hide that or we get
- 1:51:13rid of it or we try and audition something in.
- 1:51:15So we may do one take where that thing, you know, we've got that
- 1:51:19thing showing up on a volume or it's, you know, or, or we're
- 1:51:22doing, you know, live compositing like Simon Cam or
- 1:51:25something like that. And you can have an actor,
- 1:51:31they're coming to get us, but. They're coming for you, Jim.
- 1:51:33You. Guys are you guys keep talking
- 1:51:35about others future tech we're coming to get?
- 1:51:37You AI companies are here. Yeah, yeah, they're, well,
- 1:51:40they're all listening. If you guys didn't hear that,
- 1:51:42the police are coming. Yes, yes, they don't like what
- 1:51:45we're talking. We're we're giving the secrets
- 1:51:47away. But yeah, like it's it's just
- 1:51:52really fast to to iterate ideas and myself as long as as lots of
- 1:52:00people, especially lots of people that have been coming to
- 1:52:02Sigraph for a long time, have been chasing real time for a
- 1:52:05very long time. And the progression of the
- 1:52:07promise of like, you know, not waiting, you know, multiple days
- 1:52:11or weeks to render out frames that then get headed into a
- 1:52:15sequence. The sequence gets edited into a
- 1:52:18reel and then we have multiple reels which make up a film.
- 1:52:23And and that's what people like is that the capability of the
- 1:52:27real time decision making of like auditioning things.
- 1:52:31And sure, you know, you can pixel peep a bit too much to
- 1:52:35where you give somebody too many decisions.
- 1:52:37And that's like, hey, you got to make a decision here is like, is
- 1:52:40it 3:00 PM or is it midnight? What are we doing here?
- 1:52:43Like so that having that vision too is also really important.
- 1:52:47To be like you mentioned, different people coming to the
- 1:52:51engine, doing different things like Dead Maus coming for his
- 1:52:54music shows. We have the OPS, we have
- 1:52:56Stuntman's. If you have the vision and you
- 1:52:59have the foundation, you go to the engine, you can get what you
- 1:53:02want. You experiment, you will fail
- 1:53:04sometimes, but you will get what you want out of the engine.
- 1:53:07But. You're learning the whole.
- 1:53:08Time you're learning. You have to be learning.
- 1:53:09You have to be learning all the time, all the time.
- 1:53:12That's very true. Jim, is there anything exciting
- 1:53:14you're working on now that we should be looking forward to?
- 1:53:17There will be stuff. There's a couple things I can't
- 1:53:19talk about. Every single time, man, every
- 1:53:22single time we do this, we're going to talk to him about it
- 1:53:24off camera and then we'll put in the caption.
- 1:53:26I can say I mean recent stuff that I have worked on within
- 1:53:30like the past year. Obviously I got to work on the
- 1:53:33second film with Kevin Bailey and Don Burgess and Bob Zemeckis
- 1:53:37which was the Tom Hanks Robin Wright film here.
- 1:53:39Nice finished up a show, a feature film and series with
- 1:53:45Amazon. One was with called You're
- 1:53:48Cordially Invited, which was like a ROM com and more of a
- 1:53:56graphic novel that got turned into a series coming on an
- 1:54:00Amazon called Criminal. So that was an awesome one.
- 1:54:04Had a really. Stellar.
- 1:54:05Is it out yet or? No, this year TBD, this year not
- 1:54:10release date yet. OK.
- 1:54:11But you think this year or TBD? TBD.
- 1:54:14TBD. TBD.
- 1:54:15Is NDA OK? TBD.
- 1:54:17I don't even know the release date yet, but that was another
- 1:54:21fun one when we used a lot of virtual production on that.
- 1:54:25And then working with a lot of different teams who were new to
- 1:54:29both using Unreal Engine for previz and then obviously using
- 1:54:32virtual production techniques for for do things.
- 1:54:36Because that's the other thing we didn't talk about like, but
- 1:54:40you don't have to just work on sci-fi for virtual production.
- 1:54:44Or real engine? It doesn't always have to be a
- 1:54:46guy in a shiny helmet in the. Star Wars.
- 1:54:48Universe, it does really well for world building and
- 1:54:52fantastico, but it's like, you know, I've used virtual
- 1:54:56production and and real time tools for everything from ROM
- 1:54:59coms to kids shows to drama to, you know, even documentaries
- 1:55:05like there's, there's and that's the thing.
- 1:55:07It's, it's not just for like anyone genre.
- 1:55:10It's not for just LED. It's not just for games or
- 1:55:14anything like that. So it's like, you know, just
- 1:55:16like dispel the myths here and go out there and experiment.
- 1:55:20And that's the biggest thing you're gonna find out to where,
- 1:55:24where these different tools and platforms and approaches to
- 1:55:28whatever it is that you're working on.
- 1:55:30You're gonna find the sweet spot and then you know, you're gonna
- 1:55:34hone into that cause everybody has their favorites, you know,
- 1:55:37So that's the biggest thing is like getting out there
- 1:55:39experimenting with the stuff. And I know we've talked a lot
- 1:55:43about engine today. We're not biased.
- 1:55:47No, we're not. Biased, but but it but it just
- 1:55:50happens to be a tool and a platform and an ecosystem and
- 1:55:54also like just working with the team at Epic over the years.
- 1:56:00Like the first time I got to work with Unreal Engine was
- 1:56:04because I worked on a project with Epic and you know, I'm
- 1:56:08sure. They're not biased at all.
- 1:56:10No, but like I said, back to the break from thing of like looking
- 1:56:13at what's on the horizon for, for different things and at
- 1:56:16look, there is probably the, you know, the Tim Sweeney out there
- 1:56:20maybe riding the next thing that has passed real time, you know,
- 1:56:24or there is that, you know, developer, vibe, coder, whatever
- 1:56:28that role is of the future, you know, but there's so many
- 1:56:32acronyms of like, you know, it's it's like even even today, I
- 1:56:36say, you know, the prompt artists or the prompt engineers
- 1:56:39already outdated because we're having conversational AI where
- 1:56:43you're not typing anything, you're just talking to that.
- 1:56:46So sorry, prompt engineers, but yeah, that biggest things, you
- 1:56:51know, stay passionate, stay educated, you know, get out
- 1:56:56there, come to these events, you know, watch this podcast.
- 1:57:00Thank you. Because that's the most
- 1:57:02important part. That's like if Jim has said
- 1:57:05anything that holds the most amount of way today, we're going
- 1:57:09to put this in the intro and then we're going to repeat it
- 1:57:11time and time again. Oh, this is going to be there.
- 1:57:13Yeah, this will be on the short. Yeah, this will be on the short.
- 1:57:15Watch a podcast by Jim sponsored by Jim.
- 1:57:18Sponsored by me. But I just wanted to say thank
- 1:57:21you. I'm so glad we got to do this in
- 1:57:24person finally. Not going to be the last one.
- 1:57:27We're going to do this again. And I know by the next time we
- 1:57:29do this, there's probably a whole lot of new technologies
- 1:57:32that we have to talk about. I mean, we're at 5-6 now on a
- 1:57:36real engine, yes. We're not going to be too far
- 1:57:38off from 5 seven. No, definitely no.
- 1:57:40No one has comment. We don't.
- 1:57:41Know 6 is coming next year, right?
- 1:57:43No, no, no, no, no. Two years.
- 1:57:46We're still. We're still.
- 1:57:47Yeah, we look at, we look at the engine drops because it did,
- 1:57:52they did scale back, you know, and they, you know, Epic did say
- 1:57:55that the core focus for them is to go to more concentrated drops
- 1:58:02throughout the year, updates throughout the year, just not on
- 1:58:04that. And core focusing on things of
- 1:58:07seeing like how much time they've put into animation
- 1:58:10because like virtual production has matured a lot.
- 1:58:12You know, there it isn't a lot of that tooling is, is in the
- 1:58:16engine. And you know, you know, the
- 1:58:17teams with like, you know, Ryan and Alejandro and Thomas and Rod
- 1:58:22and everybody on those on those teams is now they can go off and
- 1:58:27solve some other problems that we have to do with like, you
- 1:58:29know, retargeting and control rig and mocap manager and, and
- 1:58:33all that stuff. So it's like we could do a
- 1:58:35podcast alone on all the updates in 5-6 and it's that's a big
- 1:58:39thing is like, yeah, go out there because all the updates
- 1:58:44just started hitting for on real fast Orlando.
- 1:58:47So when my talk comes out, you should all watch it.
- 1:58:51Of course there's a There's a lot of good.
- 1:58:53Ones you know when it's going to be released.
- 1:58:55I think my talk should be imminently out, like maybe right
- 1:58:59after this one goes. Live OK, we're going to put the
- 1:59:02link in the. Description yeah so if you go on
- 1:59:04the epic YouTube page you could go to any of the even the
- 1:59:09previous some real fast and then you see all the talks that get
- 1:59:12updated after that so like on real fast Orlando was in June
- 1:59:17still a lot of work to edit all there's lots of talks but like
- 1:59:22that all all the internal teams that epics that did their talks
- 1:59:25to so like all the stuff on metahuman animator updates
- 1:59:28because. Really.
- 1:59:28Exciting stuff with a big yeah, so go watch it all.
- 1:59:33I mean, it'll keep you busy for a while.
- 1:59:34Yes, there's a lot that you can watch and it's always helpful to
- 1:59:37watch them because you get to know about the updates before
- 1:59:39they release. Prepare yourself.
- 1:59:41Got any projects going on? Sneak peeks.
- 1:59:43Yeah, of course. But like I said, it's been a
- 1:59:46pleasure talking to you once again.
- 1:59:47I hope you have a great time at SIGGRAPH.
- 1:59:49We're going to be catching up time and time again during the
- 1:59:51next we're. All sneaking away to the epic
- 1:59:54parties. Yes, yes.
- 1:59:55You're gonna go there now? Yes, but thank you so much.
- 1:59:58I think it was a lot of value, especially people who are
- 2:00:01starting out, especially for people who are a bit lost,
- 2:00:03especially with all the technology advancement that we
- 2:00:05have today. I think this podcast will help a
- 2:00:08lot of artists, so thank you. Yeah, just get fueled up with a
- 2:00:12lot of a lot of good espresso and hit the engine.
- 2:00:16Yeah, So get out there. I love that.
- 2:00:17Thank you guys for watching and listening as well.
- 2:00:19See you guys in the next episode.
- 2:00:21Thank you guys so much for watching and listening to
- 2:00:24another episode of the Bad Decisions Podcast.
- 2:00:26Our goal is to bring this conversation in front of
- 2:00:29millions of people, and therefore we have one little
- 2:00:32request from you because if this conversation helped you, then
- 2:00:36you can help other people hear it too.
- 2:00:38The way you can help is by leaving a five star review on
- 2:00:41Spotify and Apple Podcasts and follow us on Spotify and
- 2:00:45YouTube. That's it.
- 2:00:46We'll see you guys next week on Tuesday.
- 2:00:49Thank you.