Global VFX Challenges: Industry Insights from 20-Year Veteran Joseph Bell
The business of visual effects is constantly evolving, and right now we're witnessing what 20-year industry veteran Joseph Bell calls a "market correction" rather than just post-strike recovery. In this eye-opening conversation, Joseph draws on his experience at ILM, Lucasfilm, FuseFX, and Technicolor to provide a nuanced perspective on where the industry stands and where it's heading.
That unprecedented content boom of 2022? It wasn't sustainable. Major studios like Disney have scaled back production by roughly 40%, and while that feels like a dramatic drop, Joseph explains we're actually returning to pre-pandemic production levels. Meanwhile, the global landscape of VFX production continues to shift dramatically, with tax incentives driving work to Australia, the UK, and France, while places like Quebec face devastating cuts after incentive reductions.
Joseph's VFX World Atlas project (available free at vfxatlas.com) reveals fascinating data points, like how 92% of rotoscoping work now happens in India, along with 50% of creature technical director positions. For independent filmmakers, this changing landscape creates both challenges and opportunities.
The conversation tackles AI's role in visual effects with refreshing clarity - distinguishing between generative AI's limitations for "final pixel" work and machine learning's practical applications in accelerating specific pipeline tasks. As Joseph notes, directors like James Cameron aren't looking for tools that get them "95% there" - they need precise control over every element.
Whether you're planning your first VFX-heavy project or navigating industry shifts as a professional, Joseph's advice to "make friends with VFX early" could save your production thousands. A VFX supervisor who spots potential issues before shooting typically costs the equivalent of just 1-3 VFX shots - making early consultation one of the smartest investments for any production.
Ready to navigate the evolving world of visual effects? Connect with Joseph Bell on LinkedIn to learn more about his consulting work and industry insights.
Transcript
Paul DeNigris (00:00)
We're talking about the business of visual effects with 20 year studio veteran, Emmy nominee and business consultant, Joseph Bell on this episode of VFX for Indies
Welcome to VFX for Indies, a podcast about the intersection of visual effects and independent filmmaking. I'm your host, Paul De Negres, VFX artist, filmmaker, and CEO of boutique visual effects shop, Foxtrot X-Ray. With me today is Joseph Bell, and I'm just going to read directly from his bio on his website, is hdriconsulting.com. Joseph Bell is a unique voice in VFX animation and post-production, equally at home, in the boardroom, and on the production floor.
His track record of delivering enduring results in the industry spans a diverse range of business situations over the past 20 years. Joseph has worked at some of the most respected names in the field, including ILM Lucasfilm, FuseFX, The Mill, Technicolor. He has built and led award-winning teams in feature films, television
well as in experience and branded content. He's also the creator of the VFX World Atlas. Welcome to the podcast, Joseph.
Joseph Bell (01:22)
Thanks Paul, happy to be here.
Paul DeNigris (01:24)
On this show, our goal is to bring on guests like Joseph to share their insights into visual effects and film production without going too deep into the weeds of tech talk with the goal of educating filmmakers who are relatively new to using VFX. If you like what we're doing here, please like and subscribe to stay updated as we release new episodes. And you can find our back catalog of episodes at VFX for indies.com. So let's get started. Joseph, why don't you tell us a bit about yourself.
and your background, your education, what brought you to your career in VFX?
Joseph Bell (01:53)
well, I always wanted to do this. Visual effects was always the thing I had my sights set on growing up in the UK, watching Ridley Scott movies and, you know, famous things from the 80s and 90s. I was like, I want to go do that. ⁓ so eventually I wound my way out to Los Angeles to go to film school. And while I was still in school, I ended up working in a visual effects producing job, ⁓ which I wrestled in around classes and, kind of got my start in VFX that way. It's a lot of fun. It led me to work at several, as you mentioned.
major studios, lot of different types of work, different teams, collaborating with teams all over the world. More recently, I've taken an interest in the industry as an industry, the sort of big picture global view of it and how businesses run within that industry, which is key when you're steering a visual effects studio through the ups and downs of the film industry. So that's where I am these days.
Paul DeNigris (02:46)
Yeah, so what prompted that shift into consulting and being more on the business side of things? And what specifically does your company, HDRI Consulting, do?
Joseph Bell (02:55)
I like to, well, I like it all to be honest with you. ⁓ but you know, I've spent plenty of time as an executive working out of the effects studio embedded, building the team and all the rest of it. also get requests from investors and people who are new to the industry, people starting businesses for sort of more, more general tailored help with whatever it is they need to do in the moment. So I don't like to think of myself as a consultant in a more advisor perhaps. ⁓ but you know, I feel a lot of questions from people.
who are looking to grow businesses, who are looking to buy businesses, who are trying to understand what's going on, you know, maybe a little bit outside of the day to day of their own company.
Paul DeNigris (03:35)
So obviously we're facing some real challenges in the visual effects industry right now. ⁓ What's your read on the situation? Like what are the biggest challenges that the industry is facing?
Joseph Bell (03:42)
Mm-hmm.
It's a market correction, I think is the way I'd look at it. And I think we all know that 2022, which was this sort of spectacular peak in demand for content, it's not really a sustainable place for any part of the film industry to be. You know, I think we saw a bit of Silicon Valley thinking with these big streaming platforms, fighting to grab audiences, fighting to get attention and spending a lot of money on content to do that.
You know, around the time that the Hollywood strikes occurred, I think the strikes got a lot of sort of headlines about how it stopped film production. But at the same time as the strikes, something else was going on in the background, which was that there was suddenly this demand for streaming platforms to become more profitable. Netflix has always been there. Disney Plus and these other streaming services racing to catch up. And suddenly they turned this corner and there was a new demand to change the way it was looking financially for them.
So they pumped the brakes on making content. I believe the number is 40 % reduction for Disney in 2023. And Disney is obviously one of the biggest clients for visual effects in the world. So you've got this huge client, suddenly they're making 40 % less visual effects pretty much overnight. So that change in strategy coincided with the strikes. I think people sort of might've missed the transition there and coming out of the strikes, things didn't pick up the way a lot of people were hoping they would.
And that's because the studios and the, know, particularly the big streamers and global studios are more cautious about spending the megabucks on content. Now it's more competitive for VFX studios. Uh, the budgets are a bit lower, just less content in general. I think we're sort of stabilizing it around 75 % of the peak amount that we saw being produced in 2022, which is not bad.
You know, we're very much equivalent to where we were, say, pre-pandemic. We've got as much work now as we did then. But it feels like less because we've come down off this huge peak.
Paul DeNigris (05:47)
Right. ⁓ You know, ⁓ I feel like the studios and the streamers, ⁓ they sort of took advantage of the strikes to ⁓ kind of put the... I guess the word is blame. To kind of put the blame on labor, right? The artists and the actors who are striking to sort of put the blame on them for, here's the reason why you guys aren't getting new content or as much new content.
Joseph Bell (06:03)
the
Mm-hmm.
Paul DeNigris (06:15)
And I feel like, you know, on this podcast and just in my day to day life, when people ask me about the business and, are you guys recovering from the strikes? said, well, the strikes weren't really the problem. You know, I sort of try to advocate for, you know, this is a management problem. This wasn't a labor problem. Not to go all Marxist on the podcast.
Joseph Bell (06:27)
Mm-hmm.
Uh-huh. I hear you. mean, my tendency
is to never attribute to sort of intention what can be contributed to confusion and perhaps poor communication. So I'm not quite ready to say, you know, that was a really sharp move on their part to run interference with the strikes. think it was just, you know, lots of messaging flying around and things happening at the same time, you know, and people wanting to sort of feel positive about
Paul DeNigris (06:48)
Yeah.
Joseph Bell (07:00)
you know, as soon as this strike is over, we're going to be right back at it. And unfortunately, that didn't turn out to be the case.
Paul DeNigris (07:06)
Yeah, for sure. What's the situation on the ground in LA right now when it comes to VFX and really to film production as a whole? Because we're hearing about studio closures, we're hearing about sound stages being dormant, ⁓ crew people not having work, losing their pension and health benefits, all of that sort of stuff. So what are you seeing as somebody who's there?
Joseph Bell (07:19)
Hmm.
It's rough for LA based crew of all descriptions, not just VFX artists, as you know. LA has pretty good, I'm a data nerd and LA gathers really good statistics, meaning that, companies based there on the amount of shooting that's happening on how booked sound stages are, you know this. And so what we're seeing is this particular point in 2025, we are way, way down.
in terms of just general production activity. Will it bounce back? One of the really challenging things to interpret about the situation we've just been through is as the work comes back, it doesn't necessarily come back to where it was before. And we're seeing, I mean, it's just the reality. We're seeing really strong tax incentives kicking in, in Australia, in the UK, in France. You know, they've all kind of upped the incentives to bring big international productions.
away from the US to those countries. When I look at crewing numbers in Los Angeles, I look at the number of visual effects artists there. I think it took me by surprise. There's actually more visual effects, I beg your pardon, more animation artists remaining in Los Angeles than visual effects artists. There are still a couple of really big animation employers. I'm looking at you, Disney. I'm looking at DreamWorks, you know, who employ hundreds upon hundreds of people in LA.
versus the visual effects studios where a lot of it has already been sent to other countries and offshore. One of the concerning trends I'm watching is that there's a lot of confidence now with these big animation studios that they can do the work. Principally in Vancouver, I think is the first stop for that. So I'm watching the numbers very carefully to see now that Moana 2 is running through teams in Vancouver, not Los Angeles. Are we going to see a new wave of unfortunately
reductions and scrambling for work because of that.
Paul DeNigris (09:20)
Yeah, obviously we can't underplay the role of tax incentives and the globalization of the industry. For me, as a small boutique studio owner, the troubling things that I see are ⁓ when companies like ILM are closing offices, when companies like The Mill and Technicolor are going through their challenges.
Joseph Bell (09:41)
Mm-hmm.
Mm-hmm.
Paul DeNigris (09:49)
DNEG is reducing ⁓ their crew and all of that sort of stuff. even the stuff that's going on in Quebec, there's a lot of VFX people out of work. It's not just a California thing. It's not just a US thing. Yeah.
Joseph Bell (09:50)
Yeah.
Mm-hmm.
Yeah, absolutely. Quebec, as you know, cut back on their incentives a little bit
in a very abrupt fashion. There was a rationale for it. You know, it wasn't crazy from the perspective of the government, but the impact it's had on visual effects and animation in Quebec has been nothing short of devastating, I would say just the reduction in jobs that are available in that province.
Paul DeNigris (10:27)
Yeah, yeah, it's, it's a shame because, you know, people, people move to these locations, they move where the incentives are, they move where the work is, and their lives are there. And then all of a sudden, the plug gets pulled. And now, where do you go? There's only there's only so much room for, you know, independent VFX artists with their own, you know, their own computer and their own nuke license, right? So there's only so much work to go around. Yeah.
Joseph Bell (10:33)
Mm-hmm. Yeah.
Mm-hmm.
Yeah. And part of the thing was the incentives, as you know, is it's tied
to where people are sitting. So, you know, if I'm a UK studio, I can't claim the UK incentive outsourcing the work to some other country. Of course, the goal is to have people there. mean, the flip side of it, we're seeing in Australia at the moment where, you new incentive, very strong, very busy. And there's a fairly, I don't want to say a limited pool of talent there. There's great talent. There's big industry there.
flexing its size on the other side of the world from Europe and the US isn't quite as easy as it might be in other places. you you've got people in Montreal who are struggling to find work because things have cut back there and companies in Australia struggling to find artists because it's going bang gangbusters in that part of the world right now. So do people want to move? Australia is lovely. Let's go. But if you have a family, you have kids in school in
in Canada or somewhere, maybe it's not that easy. So these are the challenges you face.
Paul DeNigris (11:52)
So, you know, looking into your crystal ball, what are your prognostications for the future of the industry? Short term, medium term, long term, however you want to define those. Where do think we're going?
Joseph Bell (12:06)
I think it's going to take a few years for us to gradually start approaching the volume of work we were seeing a few years ago. So it feels very abrupt that so many people are out of work, much less content being made. In reality, like I say, we've been set back a few years and I think probably three to four years from now we'll be pretty close to where we were. A lot will change by then. I think we're going to continue to see
more complex visual effects and animation work going to further flung parts of the world. I think, unfortunately, that's just what we need to anticipate. There are certain jobs that are anchored in places like Los Angeles and London because they really are best done close to clients. So if you look at, the global distribution of previs artists, of course, they're sitting next to directors in cities where there's a lot of filming, where the directors and the clients are working.
It's pretty hard to do pre-vis from wherever you're far-flung place in Southeast Asia, say. One challenge for those types of roles, even they are not safe because what we're seeing from Netflix and others is a push to commission content, particularly outside of North America. If you look at commissioning trends, Europe is pretty stable, America is dropping.
And most other parts of the world are seeing more content commissioned by the global streamers. So more directors, more writers, more production in other parts of the world. So perhaps for the first time, roles like VFX supervisor, where you want to be in proximity to the client, more of those clients are going to be in more diverse places around the world. Not necessarily a bad thing from the standpoint of the content we're getting, the stories that are being told.
⁓ It's a spanner in the works for this big cluster of VFX supervisors, perhaps in London or LA whose clients are now elsewhere.
Paul DeNigris (14:07)
Yeah. And where do you think AI fits into all of this? It's a big hot button topic. Generative AI and machine learning tools.
Joseph Bell (14:17)
Happy
to speak to it. And in fact, I think you and your audience have interesting things to tell me about it. Not just, not just in one direction. The reality is that I always pick on Avatar as just, you know, that's the, you know, the top end, big budget, huge crew, highly, you know, so James Cameron's Avatar movies. AI, you can't turn to it and say, make me Avatar. You just can't. I know Jim Cameron has been
know, vocal, he's on the board of an AI company. I'm sure the technology is very exciting to him as it is to all of us. ⁓ in terms of being able to sit back and have AI contribute to what I believe the term is final pixel on a movie like that, the pixels you see on the screen in the theater on the TV at the end of the day, Gen AIs ability to create those final pixels is pretty limited at the moment. It has fantastic applications. As you know, it's things like concept design, ideation.
Paul DeNigris (15:12)
Mm-hmm.
Joseph Bell (15:12)
You know,
I want to try out ideas. want to fail quickly. want to, you know, explore, get further with a client quicker than I could before. It's great for that. When it comes down to these really polished high-end visual effects shots, you see final pixel. The place that AI is going to have the most impact in the medium term is going to be as steps in the pipeline that we can accelerate. So a shot like that, as you know, is worked on sometimes by dozens of artists.
It goes through a factory process from animation or layout animation, lighting, know, through to compositing, you know, this. So there are steps in that process that can be made a lot less painful for the artists using AI technology. Honestly, I prefer the term machine learning for that type of application. It's sort of the same when you say AI, you picture hell from, you know, 2001 machine learning. can use that, that toolkit to drastically accelerate.
certain routine tasks, even those are non-trivial. You you'd think you'd just say, hey, do my Roto for me. And certainly great advances are being made with machine learning and Roto. I don't think anyone is quite ready to say, we don't need the human eye in Roto anymore. You know, we're not there yet. The software can't do it. The types of problems that you solve as a Roto artist.
require you to understand what you're looking at, not just to be a pattern recognition device. And so, you know, even in what you might think of as sort of the most, the least subjective parts of visual effects, it's not a magic bullet for anything. When you then turn to something like Avatar, where studios and directors are paying vast amounts of money, amongst other things to have complete control over the image. And I think one theme that's going around at the moment is that
Gen AI has sort of, you know, arguably hit the quality bar in terms of being able to generate an image. But what it can't yet do is give you precise control over different elements in that image, at least to my knowledge. I haven't seen that done yet. So, you know, a director like Jim Cameron or Steven Spielberg or your indie, you know, your indie filmmakers in the audience here want to be able to say that in the background there, that's not right. My vision is that...
You know, that, let's pick on Jurassic Park, that dinosaur there, you know, should lift its tail 15 frames sooner than it does. AI can't do that yet because it doesn't understand the dinosaur as a 3D object. ⁓ It doesn't, you know, it doesn't understand the scene the way an animator or a 3D model or, whoever does. So there's certain things it's just not equipped to do yet. Now, if directors were willing to say, you know what, forget about it. It's fine.
Paul DeNigris (17:45)
Right.
Joseph Bell (18:05)
You know, and that sounds very much like the Jim Cameron we know and love, right? He just looks at it goes, no, it's probably okay. No, he doesn't. He wants precise control over every aspect of the image because it's all, you know, it's all part of the plan, the vision, it's all architected. So the tools we use working with a filmmaker like that need as an industry to give us visual effects people the ability to take the note. If we use a tool, gets us 95 % of the way there. And then we get a note and it can't take the note. We're back to square one.
Paul DeNigris (18:10)
No, he doesn't.
Right. Right. Yeah. I was just working on a piece of concept art and, know, the gen AI tool I was using put put a cloud in front of a building in the skyline. And I was like, I will put the cloud behind the building. And I, I kind of went in circles with it for a while. just did. They kept erasing the building instead of the like, it just didn't understand what I was asking it to do. And so it just became it got to the point where I was just like, all right, I guess I'll just Photoshop this myself. I've now wasted more time than it would have.
Joseph Bell (19:00)
Yeah, fair enough. There you go.
Fall back on the old reliable ways. AI is so interesting because it's a technology thing, obviously, but it's also a cultural thing, a legal thing. There's so many dimensions to it. And what we're seeing in the Hollywood studios is a lot of caution about using particularly gen AI in visual effects for Hollywood movies. No major studio wants to have a lawsuit on their hands.
Paul DeNigris (19:20)
Mm-hmm.
Joseph Bell (19:27)
next year or five or 10 years from now, because Gen AI was employed in a film and now there's questions about the training data or compromised intellectual property or whatever it is. So there's a pretty firm line. In some cases, not all, telling VFX vendors just don't use AI. As usual, that sort of blanket statement glosses over the distinction we've talked about with machine learning.
So maybe you're a visual effects studio and you're doing very efficient work for your client with machine learning that absolutely hasn't touched any public training data. You've trained it internally or you're using a narrow model or whatever it is. So the legal issue is not there. You're probably not going to shout from rooftops to your client about how you're using AI to become more efficient. So because that will set off the alarms of, well, is it stolen training data? it genuine? No, no, no, it's something completely different. But when we use the term...
Paul DeNigris (20:16)
Yeah, I mean...
Joseph Bell (20:24)
you know, suddenly, suddenly there's concern all bets are off.
Paul DeNigris (20:28)
Yeah, and that's why, you last time you and I spoke, had just published ⁓ a little AI and machine learning policy on my website because I felt like it was important, particularly for my clientele. They want bespoke, handcrafted, unique art to support their storytelling. That's why they're independent filmmakers, right? They are not looking for machines to solve the situation for them. They want other artists, other...
Joseph Bell (20:35)
Mm-hmm. That's right.
Paul DeNigris (20:56)
human hands to be part of the process. But I've also said, yeah, I'm going to use machine learning Roto tools. know, if there's a tool that can speed this up and make sure that your limited budget as a filmmaker is being applied to creative things instead of paying for, you know, the drudgery of Roto. Like if we can reduce the drudgery of Roto by 50 %, then that 50 % can now
Joseph Bell (20:56)
Mm-hmm. Yeah.
and
Yes.
Mm-hmm. Yes.
Paul DeNigris (21:25)
go towards final pixels that may
actually make a difference for the story you want to tell. And so it is a very narrow use case. And it is we're being very cautious about the tools we use, who they're created by, how they're deployed, all of that sort of stuff. And again, the tools are great, but they're not flawless. They're not perfect. ⁓
Joseph Bell (21:47)
Yeah. It's a constantly
moving bar of sophistication and quality, as you know. I mean, one thing we laugh about in visual effects is always being asked to push the envelope, always make it bigger, more detailed, more real. I think we've yet to really see a technology in visual effects that let us shrink the crew dramatically overnight, because it's always accompanied with great, that's easy for you now. Can you do this next thing? Can you take it to the, can it be 4K, 6K? ⁓
stereo, whatever it is. you know, so there's always a way to, I mean, really the Hollywood studios are in the business of always being at that upper envelope of spectacle, of spend, budget, excitement, marketing. You know, that's the game. And we all know, and we're thrilled about lower budget movies that break out and do tremendously well.
Godzilla Minus One winning the Oscar was a ⁓ minuscule visual effects budget. that's all, it's all fantastic. In the general day to day of things and the Harry Potters and Jurassic Parks of the world, the bread and butter of the Hollywood studio business. They're not necessarily, I mean, they're cost conscious, but they're not looking for cheap. They're looking for that sure bet spectacle, high quality thing that's gonna get everyone around the world to wanna go see it.
Paul DeNigris (23:11)
Right, and you know, the Godzilla Minus Ones and the Ex Machinas and the films that I work on, they all are the beneficiary of the R &D budgets of the Avatars and the Harry Potters and the Marvel movies and you know, Lord of the Rings, right? We're using, I have on my little MacBook here, Nuke, which is by and large built on the innovations that were created for Avatar, for Lord of the Rings, right? These are tools that they,
Joseph Bell (23:16)
Mm-hmm.
Mm-hmm. Yeah.
Mm-hmm. Mm-hmm.
Mm-hmm. Yep.
Paul DeNigris (23:40)
they do trickle down. ⁓ So yeah, any and all high level technological exploration by those students, by the folks who have the deep pockets who make the billion dollar movies, please, we want them because five years from now, it's gonna be part of Adobe Creative Cloud or you know, and yeah.
Joseph Bell (23:47)
Hmm.
Yeah, that's an important point. And
of course, there's no sort of central council of, you know, film technology advancement for the good of filmmaking mankind. You know, there's no structural plan to it. All of this innovation comes from, Jim Cameron has a new movie and it's going to need the next level. Okay, well, you know, the budget of that movie is going to fund a certain amount of R &D, a certain amount of breakthroughs and developments that, you know, will go on to the next movie. So it happens very sort of organically.
The other thing that's very exciting to your point is the idea that these tools will let us tell stories we couldn't tell before on budgets that would have been unattainable previously. I don't have the details of Here, the Tom Hanks movie, but I understand that they used AI tools to help with the de-aging process. And that's a kind of example where you kind of feel like that shouldn't cost a gazillion dollars to de-age actors. But then again, if the whole
conceit of the film, which I apologize I haven't seen, you gotta tell me, is that the actors are their younger selves for large portions of it. That's really gonna add up. mean, one of the things I think that sometimes takes indie filmmakers by surprise is they write something into a script that doesn't seem like it's such a big deal, and then it inadvertently appears in 220 shots. Then you get into post-production and you learn that it's gonna cost.
$500 a shot to change that to what you thought it was. And you multiply that by 220 and now you have a situation on your hands. So it's the idea that we can reach for storytelling tools like de-aging more readily and it doesn't have to be a major studio that bankrolls something like that. It's exciting, I think, for all levels of filmmaker.
Paul DeNigris (25:25)
Mm-hmm.
Yeah, I mean, the deep fake stuff that's being done on YouTube, you know, where they're sticking Arnold Schwarzenegger's face on Will Ferrell's body or, whatever, right? Like, we're getting really amazing results with consumer tools, and it's because of those R &D cycles, which is great. So.
Joseph Bell (25:50)
Die.
Mm-hmm. Yeah.
Mm-hmm.
Again,
the sort of Hollywood version of that, you know, I think there've been music videos where you know, face swapping of celebrities and things like this. To my knowledge, that's always assisted with traditional VFX work with traditional VFX artists. So again, it's not like, you know, a high end post production studio goes into the back room and they hit a button that puts, I don't know, know, celebrity A's head on celebrity B's body and they're done.
No, there's a cleanup process, there's compositing, there's looking at the lighting and there's fixing the things the AI didn't quite get right. Will it always be like that? Maybe not. But right now we're still in the phase where, you know, you want the human pilot to land the plane.
Paul DeNigris (26:52)
⁓ So you're talking about the VFX ⁓ budget sneaking up on you, the number of shots sneaking up on you as a producer. one of my consistent questions on this show for my guests is in line with the show's mission, what is the piece of advice you would share with producers who aren't experienced with VFX if they want to use VFX on a project? And you sort of touched on that, if there's more to that, I'd love to hear.
Joseph Bell (26:58)
Mm-hmm. yes.
Mm-hmm.
my
goodness, thank you for the opportunity to speak to this. Thank you, thank you, thank you. I would say that it become friends with VFX early. You know, as soon as you have an inkling that you're going to need VFX to help you solve a problem, get in touch with the VFX supervisor, get in touch with your VFX producer friend and just talk to them about it. All of these people are very skilled at reading scripts that are a long way from being shot and they will spot things that perhaps you won't.
I used to do a lot of bidding when I worked at Cosa VFX. would bid lots of TV shows, you know, and I'd get to the scene on the rooftop and I'd be like, they're not going to be able to shoot this on a real rooftop. I'm going to call it, hey, are you going to have a location for this? you're not. Okay. Glad we talked about it. I'll just add 50 green screenshots to the bid right now. And so as you go through it, yeah, the car crash. You know, they're going to have a spectacular stunt. It's going to look great. But the piece of glass that lands.
right in front of her in the script, they're probably not going to be able to get that on the day. So you sort of learn what they're not going to be able to get on the day. Whenever we used to bid work with practical puppets, which was an occasional thing, it doesn't happen as much, but people would say, no, no, no, don't worry. It's going to be, it's all going to be the puppet. And they're like, okay, you get into post-production and whatever it is, the puppet didn't quite have the facial expression you wanted or didn't quite look in the...
direction of the person who was speaking in that shot, CGI is on the puppet. again, by seeing these examples over and over again, you kind of learn to catch them. Make friends with a visual effects person, get them in early. Obviously there's nothing to be afraid of. Most visual effects people are not looking to inflate your budget. In fact, I would be hard pressed to find anyone who would deliberately do that. Sometimes...
they will assume a more difficult thing is needed than a simple thing. And that's a point where you can rein them in. That's a point where you say, ⁓ no, no, that's not going to be CG. You know, it happens largely off screen. You know, we don't need to build that whole. So you can you can sort of rein in the scope, but nine times out of 10, they're sitting there figuring out how to do it most efficiently, how to build the least in CG to your vision. And, know, when things I think
sort of get out of hand is where filmmakers are asking for perhaps more than they realize. It's like, well, let's just build the whole environment. Wow, okay, well, making that look real is gonna be expensive and difficult. ⁓ If we just had to enhance outside the window or something more modest like that, your budget would be a whole order of magnitude lower. So I think one thing to be aware of is that...
Too many, I think, client-facing visual effects people have been conditioned not to push back on clients. And even at the high, high level where you're working with, you know, the Michael Bayes and Jim Camerons of this world, that can lead to issues. know, Michael wants to shoot it a certain way. He's going to shoot it a certain way. You let him. And then you go back and you figure it out. But when the budget is tighter, when the, you know, the timeframes, when it's perhaps, you know, a smaller movie,
Paul DeNigris (30:15)
Mm-hmm.
Brett.
Joseph Bell (30:32)
going back and figuring it out may not be the option, or there may be fewer options. And so I would say it's important that your visual effects people feel empowered to speak up and say, oh my goodness, if you move the camera three feet to the left, you're gonna save $100,000. If they don't feel empowered to do that, you're not gonna get the good information. Then you can at least make an informed choice.
Paul DeNigris (30:47)
Yeah.
Absolutely. I always tell people it's less expensive to have me on set for a day than it is to hire me and my entire crew for a week to fix the thing that I would have caught on set.
Joseph Bell (31:00)
Thanks.
Mm-hmm.
Hopefully you
reinforce that on every episode of the show and eventually someone will do that.
Paul DeNigris (31:09)
Yeah.
Yeah, that's that's the idea. ⁓ Yeah, right. Like right now, literally this morning, we are bidding on a short film where they inadvertently realized, we need to replace this sign in the background in 22 shots or something like that. It's a short film. They don't really have the budget for this. And had ⁓ somebody been on set who was.
Joseph Bell (31:23)
Yeah.
Mm-hmm.
Mm-hmm. Sure.
Paul DeNigris (31:36)
VFX savvy, not even a VFX supervisor at that point, right? But somebody who was aware of VFX, if they were like, hey, if there's even a thought that we have to paint that out, we need to shoot this a different way. You know?
Joseph Bell (31:40)
Mm-hmm.
Mm-hmm.
I'm going to generalise
grossly and say that the cost of having a visual effects supervisor on set for the day is equivalent to the cost of doing, in sort of traditional VFX facility terms, maybe one to three shots. So if you can avoid three unnecessary VFX shots because your VFX supe was on site, they've paid for themselves.
Paul DeNigris (32:08)
Right. Right. that's, mean, even if I'm coming into a project in post, I always see my job as being the guy who figures out how to say yes, but here's how we get there in a way that doesn't absolutely destroy your budget. know, how do we, how do we solve your problem? ⁓ you know, an example I always use, I had a client, they wanted to do a, they were doing a Christmas movie and they, they were shooting outside of a building and they couldn't get inside to put
Joseph Bell (32:13)
Mm-hmm.
Mm-hmm.
Paul DeNigris (32:38)
Christmas decorations and so they were like, well, let's just frost the glass so we can't see in. And I said, well, if we do that, we now are frosting the glass in, you know, two dozen shots as these characters have a conversation. How about instead we just in the two shot where we can actually see in the building, we put some Christmas decorations. And then in the over the singles, right, we can put like, you know, the bough of a Christmas tree.
Joseph Bell (32:42)
Mm-hmm.
Mm-hmm.
Paul DeNigris (33:06)
that we've established in the two shot, just kind of stick that in the frame in a way that doesn't create any roto. And over here, we stick a Christmas light in in a way that doesn't create any roto. Now we've done 24 really simple VFX shots for the same amount of money, actually less because I've reduced all the roto that I need to do. And you get Christmas decorations, which is what you wanted to begin with. Right. And they kind of like flipped out because it was not an obvious solution for them. And it was like...
Joseph Bell (33:10)
Mm-hmm. Mm-hmm. Mm-hmm. Mm-hmm.
Mm-hmm.
Yep. Yep.
Mm-hmm. Yep.
Mm-hmm.
Paul DeNigris (33:34)
You know, they think, it's just, yeah, it's just gonna be easy. Let's just frost the glass. No, the things that clients generally come to me and say, this is gonna be really easy. They don't really have a concept of how hard it is.
Joseph Bell (33:38)
you
Nah.
And the problem was that, which you by the way also see on big studio productions is the cost of currently, I should say, with the technology we have now. The cost of doing that work seems disproportionately high. The classic example of that is a reflection removal. Okay. So you have a dolly shot and you're going in front of the front of a glass building and you see your entire frickin' crew in the window, you know, with everything. So, but you know, that should be easy, right? You should.
be able to paint that out? Well, no, think about it. Because there's something semi sort of behind the reflection that we have to rebuild. And there's all kinds of parallaxing. there's you know, so it's actually that can be very time consuming, expensive to remove a reflection.
Paul DeNigris (34:32)
Yeah. And then the actors that we're focused
on are crossing in front of the area that we have to paint out. So now there's all of this roto, right? And if the camera's moving in multiple axes, it becomes a 3D camera track and projection. Yeah. And it's.
Joseph Bell (34:37)
Yeah, exactly.
Yeah. So I've
done, I've done TV shows, you know, network or streaming TV shows where the most expensive shot in the show, even if it's a modest visual effects show, I'm not talking like Andor or something, but you know, it's a show that is going to spend, you know, hundred, 150,000 on visual effects for the 45 minutes of the hour. Their most expensive shot is the reflection removal and even experienced clients and post supervisors have a hard time with that. It's like, but you
Paul DeNigris (35:05)
Mm-hmm.
Joseph Bell (35:12)
blew up a building and it costs this and you removed a reflection and it costs more. And it's like, yes, that's because of the actual amount of work we had to do to do that. Now, if, you know, back to gen AI, if AI tools allow us to accelerate that process, I don't think anyone's going to be upset about that, including the artists that had to sit there and 3d track a reflection.
Paul DeNigris (35:28)
No. No.
No, no, we'd rather spend that time and energy making that building blowing up, the coolest building blowing up that's ever blown up. Yeah, yeah, for sure. So tell me a little bit about the VFX World Atlas. What is it and who is it for? I know this is one of your projects. ⁓
Joseph Bell (35:41)
Yeah, tell the story. Exactly.
So for anyone who doesn't know, probably quite a few of you, you can go to VFXatlas.com and download for free this report I made on the shape of the global VFX industry. So it's been really hard to get data, hard facts such as they exist on the global VFX industry. There are certain countries where they do lots of surveys or they track spending on VFX because of tax incentives, how effective are they?
You know, Quebec has great data. The UK has great data. Up until now, there's been no sort of consistent way to compare data across regions or really to get a big global overview of what the VFX industry looks like. So about a year ago, I was reading one of my favorite VFX magazines and they had a number in there for the size of the VFX industry, dollar amount. was looking at it, it's like, wow, that seems really, really big.
I'm not sure, I feel like we would need twice as many visual effects artists in the world to get that number. So I did a little bit of research and I found where the person who had written the article had Googled for the size of the VFX industry and a report popped up. It was a report written by someone who was not a visual effects industry insider. ⁓ And I feel like ⁓ that report probably, know, construed the whole of Autodesk and Adobe's revenue as visual effects revenue, which of course is a...
a huge mischaracterization of the VFX services market, which is really what we're supposed to be looking at. So I just realized that not only do we not have good data, but the data there is being prepared by people who don't know the industry. So I set out on, I have a very high tolerance for being detail oriented and data, a little bit OCD. I can get through a big spreadsheet. I...
made a project of scooping up all the data I could find on the internet about people who work in visual effects around the world and bringing that together into a single publication, an atlas, where you can look at a map and you can see where visual effects people are. So it's kind of nerdy. Check it out if you're at all into that. And there's lots of facts and figures in there. ⁓ I don't think anyone was sort of blown away by what, because we all kind of knew. You kind of know there's a lot of VFX artists in India.
I mean, you know that, but how many are there? Well, it turns out that 92 % of roto paint artists are in India, worldwide. If I look for every roto paint artist I can find in Canada, the UK, Australia, wherever it is, 92 % are in India. About 50 % of creature technical directors are in India these days. So it's not just the less subjective work that's getting farmed out.
Paul DeNigris (38:16)
Yes.
Mm-hmm.
Joseph Bell (38:44)
And so, you know, picture starts to emerge where the work's being done. We can see that visual effects in New York skews heavily towards commercials work. It's a big advertising marketing center, right? It's not surprising, but again, we can now put a number on that. can say, you know, of the, so many people who are doing visual effects in New York, this many are doing animated films, this many are doing general VFX, and these guys are doing commercials and design.
Paul DeNigris (38:56)
Yeah, not surprising.
Joseph Bell (39:09)
In terms of the audience for it, think it's really, this is such a basic thing that, mean, I really want to say it's for anyone who's remotely curious about the VFX industry, because again, without wanting to sort of overstate it, we really didn't have this satellite view of the industry before last year. I mean, I'd be hard pressed, maybe in the competitive business affairs office of some large VFX studio somewhere, they have analysts doing this stuff, but for us out in public.
There was no way you could go to see a map of the visual effects industry or see how big the different hubs were relative to each other. So it's something you feel like that's obvious. You should know that. Well, you know, now, now we can. So I'm finding a lot of interest in it from policy makers around the world, trying to see how their jurisdiction is doing with attracting VFX work. Are they growing or shrinking, someone else growing or shrinking?
The data set behind the Atlas is actually quite detailed when it comes to companies. I don't see the Atlas as the place to really get into the nuts and bolts of how any specific company does things. But we certainly have companies that could gain competitive edge by studying the trends, seeing where people are. Even on the studio side, I think if you suddenly had a big show you needed to do in South Africa or South America or wherever it is.
Paul DeNigris (40:13)
Mm.
Joseph Bell (40:33)
You know studios down there, but do you know how many animators they have in Johannesburg right now? You know, is that the place if you send an animated feature down? I mean, I'm sure they have plenty, don't get me wrong, but you know, it can also give you sort of Intel on the lay of the land if you're looking to work internationally.
Paul DeNigris (40:48)
Yeah, that's great. the data real time? it updated periodically? How is that working?
Joseph Bell (40:55)
So
the Atlas itself is going to be an annual publication. It's a huge amount of work to do it and it's free. So can't do it every five minutes. But the data set itself is refreshed every three months. And let me tell you, it's a steep hill to climb. since we launched the first Atlas last year, which was purely VFX, and that had 55,000 entries in it.
people working in VFX around the world. Now, just to be clear, it's not a crew directory. People aren't in there individually. They're all gathered up into statistics about how many people work in London and what types of compositors and things like this. But 55,000 data points. We're going to launch a new edition this year in July, and it's going to have north of 120,000 data points. I think we're probably going to end up around 150. It now includes a pretty good slice of the animation industry. Companies that are working
exclusively on animated content that don't also do VFX. ⁓ So it's a much bigger data set. So to clean that up every three months, we do a fresh data pull top to bottom, one data point at a time, and then we cross reference it against the previous quarter's data pull see what's changed, see what needs to be tidied up. And I like to say that it's not that the data is special. Look, you can go on the internet and get data.
All the data we use is publicly available for free. You can find it too. The thing that makes it special is that really quite a lot of effort goes into cleaning it up, screening out dud entries. I'd say 20 % of the data points that come back don't belong in our data set. And I haven't been able to train AI to spot those. I have to spot those. You I have to say, you know, if you say you live in Chile and your job description is farmer
Paul DeNigris (42:36)
Sure.
Joseph Bell (42:44)
then you probably do not work at Industrial Light and Magic. Probably. ⁓ Maybe you're a fan of Industrial Light and Magic. I don't know. But anyway, for the purposes of our study, I think I'm going to drop that data point from the data set. Something like that. Anyway, so it sort of goes to show. first of all, I do it with care and a certain amount of love because of the passion and respect I have for the industry. The last thing I want to do is put out shaky data into an industry I really care about.
Paul DeNigris (42:56)
Industrial llamas and I don't know what the M would stand for.
Joseph Bell (43:13)
⁓ But beyond that, you know, with 20 years under my belt, you know, and it's not just me. mean, I have a small team, but, you know, I can help them identify that yes, that, you know, that data point doesn't belong. It's going to skew our data and really edit it down to the core of what we really have.
Paul DeNigris (43:30)
And where is this, where are you pulling the data from? I mean, you don't have to go into specifics, but.
Joseph Bell (43:32)
It all comes from, it comes from
public websites on the internet. You know, so you can, you can browse around again. I don't have access to anything you don't have access to. It's really, you know, the, the, effort to collate it.
Paul DeNigris (43:44)
So if you were a VFX supervisor slash Nuke compositor in Phoenix, Arizona, how would you, how would I be represented in the Atlas?
Joseph Bell (43:52)
Well,
you would probably find a dot on a map that says Phoenix. The size of that dot represents the number of people working in that location. Again, there's a sheer volume of data. The Atlas doesn't give you every single city. mean, we talk about particular hubs, but you can always email me and say, hey Joseph, what you got for Phoenix? It's very easy for me to say, I've got this many people in Phoenix. It's not everyone. This is just the...
Paul DeNigris (43:58)
Thank
Joseph Bell (44:21)
the people I have in my dataset and you know, this many of them happen to be compositors.
Paul DeNigris (44:27)
Right on. Hopefully I know all of them.
Joseph Bell (44:33)
You probably do. And one of the things
that's fun about working on this data for me, and you'd find the same is, you know, at this point in my career, I've probably personally worked alongside 3000, 3500 visual effects artists around the world. So as I go through the data, you know, I don't sure who should I pick on Kansas City.
You know, there's a dot on my map in Kansas city and I know that's Anthony. Hey, Anthony, how you doing, man? So, you know, within this huge data set of tens of thousands of people and, London was, you know, eight, 10,000. There's a dot on the map in Kansas city. And I know for a fact that's Anthony. He really is there.
Paul DeNigris (45:03)
You
Yeah.
Yeah, I mean, we are we are a global industry, but we are it's we're a small town in a lot of ways. Like we all know each other. And and I guarantee if you and I spent a couple of minutes, we could probably track down somebody that we multiple people that we know in common and have worked with. ⁓ Yeah, it's it's always amazing, you know, to to think about and a lot of
Joseph Bell (45:31)
yeah.
Yeah, I'm sure we could.
Paul DeNigris (45:43)
I encounter people all the time that that worked at, ⁓ you know, the same facility that I worked out, worked at out in L.A. and we maybe missed each other by, you know, a month or a week and know everybody in common from that facility. And, ⁓ you know, I know it's yeah, I get I get and I get messages on YouTube all the time. my God, you know, Rick, I worked with Rick on Star Trek, you know, whatever. Right. ⁓ So, yeah, my guests, my guests do. ⁓
Joseph Bell (45:49)
Mm-hmm. Yep.
Mm-hmm. Yeah, absolutely.
Paul DeNigris (46:12)
do get recognized by my audience. And you're right, there's a lot of people in India. I get a lot of comments on my Nuke tutorials from folks in India. I'm a little bit of a data nerd too when it comes to like who's watching my YouTube videos and stuff. And so can see that I'm getting a lot of hits from folks in India. yeah.
Joseph Bell (46:14)
I'm sure there's crossover.
Sure.
Mm-hmm. Mm-hmm.
Mm-hmm.
And fantastic work comes out of India. We all know this, but
you know, those, those are really big facilities. Like I think the biggest VFX facility in the world is in Mumbai, just by head count. A lot of people would say it's, you know, it's Framestore London or it's WETA in, in Wellington. Those are huge facilities, but the biggest, you know, single site head count in visual effects is, is I think in Mumbai and the quality of the work that comes out of it, the sophistication of it.
Paul DeNigris (46:52)
Yeah.
Joseph Bell (47:04)
You know, I would say equal to anything else out there.
Paul DeNigris (47:08)
Yeah, absolutely. So what's something that you've seen recently VFX wise that got you excited that you really enjoyed?
Joseph Bell (47:18)
wow, you're putting me on the spot here.
The scene that's coming to mind, ⁓ my goodness, I can't remember the name of it. The Chinese science fiction series that Netflix recently made, Three Body Problem, that's right. The scene where they use the monofilament fiber to cut the ship in half. Not sure if you've seen that. That's absolutely, it's a novel idea. mean, the book is very much based in hard science, I understand.
Paul DeNigris (47:32)
⁓ Three Body Problem. Yeah.
I haven't, I haven't caught up to it.
Joseph Bell (47:50)
You know, is this possible? I don't know. But they have a gag where, you know, I believe the, one of the antagonist characters has a ship he's traveling on through the Panama Canal. I think it is with all of his people and they have to break some, some chochki out of the ship without anyone finding out or whatever it is. So the way they do this is they string a monofilament five, it's actually a net of fibers across the canal.
and the ship's own mass just pushes it through those fibers. And so you see that the ship basically go through like a, what do call it, a wire grater or something like this. Yeah, including people on the ship, I should say. mean, it's done quite tastefully, but there's no getting around it. It's really, and you know, just the sort of the execution is great. The novelty of the concept is great. And it really, they spend time on it. It's a sequence that sort of ⁓
Paul DeNigris (48:27)
Cheese grater. ⁓
Yeah.
Joseph Bell (48:46)
builds momentum, it's satisfying, I it takes time for the ship to pass through the wires so you get time to enjoy the effect. At the end of it all the ship kind of beaches on the side of the waterway and kind of just sort of sloughs off itself like a sandwich or something like this. Anyway, it's really nicely done. Kudos to whoever pulled that
Paul DeNigris (49:06)
that makes me think of the movie Ghost Ship. There was a gag where a cable snapped like everybody on the ship was on was on the, you in the ballroom and the cable snapped and just cut the entire everybody in half. And so then the, you know, the salvage people are going and trying to find something on the ship years later. And it's, haunted. So it sounds like somebody saw Ghost Ship. It was like, I want to do that times a thousand.
Joseph Bell (49:19)
Yeah.
Yeah, I cleaned up the mess, I'm sure.
I haven't read the book,
but I've got a feeling it's probably described there as such. It's interesting, but again, the gag plays out over bit of time. So have people inside the ship sort of seeing this disturbance coming towards them, but they don't know what it is. You can't see the fibers, right? You just see things falling off the wall or a picture cut in half or something like this. Like, is that? Anyway, awesome.
Paul DeNigris (49:50)
I'll have to catch up with that show.
Yeah, right on. ⁓ So where can people find out more about you, what you're up to? You mentioned the, I've mentioned your website earlier. You mentioned the VFX Atlas website. If we can just reiterate those.
Joseph Bell (50:02)
Mm-hmm. Yeah. So I mean, the easiest
place to find "me" me is probably on LinkedIn. So, you know, go look me up on LinkedIn, Joseph Bell. ⁓ And I've got quite a few visual effects people who follow me there. I post interesting things I come across. And any other news with my projects or whatever tends to go through that channel. If you're interested in the VFX Atlas, again, it is free. I don't think there's anything quite like it out there in the world.
If you're a data nerd, you want to go to vfxatlas.com and download it. You do not need to sign up or give an email or again, pay anything. It's just there. HDRI Consulting is my portal for dealing with clients, companies, studios, investors. But again, LinkedIn is the way to get hold of me if you want to.
Paul DeNigris (50:47)
Well, thanks so much, Joseph. It's always a pleasure to chat with you and you're just a font of knowledge and and and just a just a super person like personable and positive guy like I just you know, you and I had a zoom chat a few months ago. I contacted you out of the blue, I think on either on your website or on LinkedIn and and you were very gracious with your time and and I appreciated it then I appreciate it now.
Joseph Bell (51:10)
Mm-hmm.
Paul DeNigris (51:16)
⁓ you're doing a great service for the, industry with, with all of your, your data wrangling and advocacy and all of that stuff. So, ⁓ so I thank you for being here and also just thank you on behalf of the industry for, everything you do.
Joseph Bell (51:29)
well, thank you. Thanks for having me on, Paul. This was such a pleasure.
Paul DeNigris (51:32)
Yeah, thanks. And that brings our episode of VFX for Indies to a close. Thanks to all of our viewers, watchers, listeners for tuning in. If you enjoyed this episode, please like, follow, subscribe, leave a comment, a review. All of those sorts of things help us get the word out about the show. For Foxtrot X-Ray, I'm Paul DeNigris. Thanks so much for being part of the VFX for Indies community.
