Check out www.VFXRates.com.
Upload The Productive Artist e-book.
Episode 144 — Soho VFX — Berj Bannayan
This is Allan McKay. Welcome to Episode 144! I’m speaking with Berj Bannayan, one of the Founders of Soho VFX. We get into so much cool stuff, like origins of Soho VFX and Berj’s background — which is more technical.
Let’s dive in!
FIRST THINGS FIRST:
I. [-1:06:35] I want to mention that there was a bit of a mix up about the free training on Energy Ball. Since then, I had to fly down to LA, to do a quick Marvel gig. So, the dates have changed.
The free training is available at www.allanmckay.com/plasma/ or www.allanmckay.com/energyball/. This is 8-hours of free training, in a span of 7 videos. We get into a lot of effects, lighting, integrating, etc. Check it out!
II. [-1:03:04] After the Launch, my website will be going live.
III. [-1:02:52] We will also be publishing a free guide on Negotiating Mindsets — FOR ARTISTS! Again, to learn about all the most recent releases, please go to www.allanmckay.com/inside/.
INTERVIEW WITH BERJ BANNAYAN
Founded in 2002 by partners Berj Bannayan, Allan Magled and Mike Mombourquette, Soho VFX began as a boutique studio. From the beginning, the studio’s specialty was the production of high quality photo-realistic visual effects. In 2005, Soho VFX landed itself a leading visual effects role: the creation of the elastic Mr. Fantastic in Marvel’s The Fantastic Four. Since then, Soho VFX has helped superheroes come to life in several other Marvel productions.
In 2007, Soho VFX expanded its facilities again in order to take on almost 200 shots for The Incredible Hulk. The studio’s success on this blockbuster project propelled it to become a world-renowned visual effects house.
Since then, Soho VFX has continued to produce top-quality visual effects, contributing to many box office hits, including: Logan, Tomb Raider, X-Men Origins: Wolverine, The Revenant, Rise of the Planet of the Apes and many more. *
In this Podcast, Berj Bannayan talks about his background in computer science and software development; the launch of Soho VFX and the lessons they’ve learned along the way — and the massive projects they’ve tackled.
Soho VFX Website: https://sohovfx.com
Soho VFX Reel: https://sohovfx.com/reels
Berj Bannayan on IMDb: https://www.imdb.com/name/nm0052294/
Soho VFX on LinkedIn: https://www.linkedin.com/in/soho-vfx-96987696/
Soho VFX on Twitter: @SohoVFX
[-1:02:08] Berj: Yup, my name is Berj Bannayan. I’m one of the owners and supervisors at Soho VFX in Toronto.
[-1:01:59] Allan: Awesome! And I’m Australian. So I’m always going to butcher people’s name. You didn’t come from an artist background. Did you always think you would work in film or what were your expectations?
Berj: Well, I’m a child of the 70s. I grew up with Star Wars. That was a big influence on me as a movie goer, as a child. But [I also thought], “Wow, this would be amazing to do [and] to be a part of!” That began in the early 80s. I was 10 years old. As I moved my expectations of career toward a university in the 90s, I thought, “Hey, this would be cool!” I could try doing some of these things. I had a computer back in the 80s and I was doing some modeling — and I discovered I was terrible at it. Absolutely awful! I had some aptitude for software development, so I started doing Computer Science. The industry was moving with computers and I remember seeing these computer-heavy effects in Jurassic Park — and I decided to do it that way.
[-1:00:24] Allan: That’s really cool! I think it’s unique too. People think that if they aren’t getting the results they want, they either work harder or quit. But for you, it was about adapting and finding what came naturally. Was there a specific area you decided to specialize in, or did you decide to go the technical route?
Berj: At the time, when I was in school, the computer animation and digital side of things was still pretty new. I started the university in 1993, I think. So Jurassic Park was still a couple of years gone at that point. Toy Story hadn’t happened yet. These digital things didn’t really exist. So I started looking at companies that made software. Back in the day, there was Alias, now part of Autodesk, and Side Effects who made Houdini. They’ve become fundamental in our industry.
[-59:12] Allan: And down the street from me.
Berj: Very close, actually. My university had a coop program where you studied for 4 months and worked for 4 months. So I ended up at Side Effects working on — at the beginning — Prisms. They were transitioning into Houdini years and so I ended up being part of the software development team. I was able to be exposed to wide swats of the industry because they had such a diverse set of clients. Then I decided to transition into production and I ended up at a visual effects facility, here in Toronto. I left that in the late 90s.
[-58:28] Allan: That’s really awesome! I’ve met a few guys at side effects. Jeff Wagner and a few others. From there, what was a big break, in terms of getting into actual films?
Berj: I think the big moment was when I started working at a company called C.O.R.E. Digital Pictures, which is sadly no longer around. I think it was in the summer of ’97. They were starting to work on the Doctor Doolittle. They didn’t have any full-time software developers on staff. I had convinced them that I would be useful, having just come from Side Effects. They were just starting to use Houdini, or what would become Houdini. So I was tasked with writing the software that would create the fur on digital rats. That was a new experience. That was the first time I was able to have an impact on the end result of a film — which was very exciting and terrifying at the same time!
[-57:12] Allan: That’s pretty awesome considering that was the beginning of the animal craze. You would involved in Spawn and The Big Hit too.
Berj: Yes, that’s right! The Big Hit was the first time I got to go to set, with my boss at the time.
[-56:48] Allan: What did you, guys, do for that? I’m just curious. I tried to watch it the other day.
Berj: Oh, man! It was the 90s. If I remember correctly, there was a scene where a character is running down the hill and there is a car after him, and the car flipped. They were supposed to do it practically. The car never flipped the way they wanted it to, so we ended doing a digital swap as it was in the air. It’s that whole beat. It was a lot back then.
[-55:58] Allan: That’s cool! That was toward the end of the movie. And what about Spawn?
Berj: We came into that movie late. I guess the studio discovered they had something interesting on their hands and they threw extra money into it and brought on more visual effects houses. We did a bunch of stuff with his bike when he’s chasing the clown, pouring some gas out of the truck. And the bike has all of these lightening effects. That was a fun project! We got a big piece of the bike. There was no cyber scanning back then, as far as I knew. We had a scanning robot arm, basically, where you put the arm point by point along the surface. I can’t remember what it was called. That was interesting.
[-54:36] Allan: It’s so interesting to watch kids knock stuff out in Zbrush these days, in a couple of hours. Back then, I used one of those. I think you had a pedal.
Berj: It was an interesting thing. Also, 2D tracking was starting to become a thing. There was a product from a company called Hammer Head. I was doing a lot of the tracking and bringing the software to transfer the data from there to Houdini and… What was the compositing package back then? I think it was called ICE. So it about bringing the tracking into the 2D or the 3D world. I always had the technology focus which was a lot of fun.
[-53:25] Allan: That’s really great! I’m curious about X-Men from your pre-Soho days.
Berj: That was fun, actually. That was one of my last projects before I left. We did the map table with all of the pins that made shapes of the city — which was fun! And that was a big software project. My boss at the time came up with a plan of how the pin tables would work, but they were doing it procedurally in Houdini and it was very slow. It needed something custom. I was able to build a piece of software that basically allowed you to simulate that pin table. You could model a shape, put it on top of the table and the pins would do a spring simulation and bounce as they hit the edge of the surface. And you could move that surface around, as the story of the map zooms in. It was a heavy asset and the rendering of it was intense. It was a slow process because the simulations took a long time. The animators had to track the table and the cameras, we had the simulations, the lighting simulation. It was an exercise in a very heavy pipeline. As it turns out, it wasn’t so much screen time, but people still ask me about that one.
[-51:11] Allan: It’s kind of like the bullet in The Matrix. Everyone wanted that effect. I was working on a commercial and the client wanted the thing from X-Men and we had a week to do it. That’s awesome! I was a big fan of that! When you launched Soho, there was a team of you. How did it come to be?
Berj: It was sort of the right time. My partner Allan Magled and I were working together. You talk a good game when you’re working for someone. We had an opportunity to go and build a VFX department at another company Soho Post and Graphics. They were a commercial effects house. We were 5-6 years into our careers. We jumped in. I believe it was winter of 2002. We had a few people with us in the beginning. We had to do everything, we had to build it from scratch. It was an eye opener. You forget the infrastructure that you take for granted. We did that for over a year, at Soho. They wanted to exit the industry and we bought the company from them and went on our own ever since. It’s been 15 years now. We just decided to jump in with both feet — and here we are.
[-48:00] Allan: What’s it like starting a new company? A lot of people when they’re an employee, they don’t think about the pressure of running a studio. Was it overwhelming to build something from the ground up?
Berj: Yeah, it’s a combination of exhilarating and terrifying at the same time. We knew how to do what we knew to do. I knew software. Allan came from animation background and moved onto Supervising. We had a couple of other guys and each person had a task, which is why I think it worked out well. Another partner of ours Mike Mombourquette was at Soho before we arrived. We absorbed him from the other company. Each of us had a job to do which made it easier than hiring out, or each of us doing everything. We managed to collect a small team and each [person] brought something to the table. In that way, we blissfully went about our business. We started building relationships with clients and studios, and building our brand. There was a lot of “ignorance is bliss” in the beginning. We just went in and did it.
[-45:29] Allan: Going in there with a naivete is sometimes better. I think that’s really great! What are some of the bigger challenges with growing a studio? I think it’s smart to grow slowly and steadily.
Berj: I think for us it was about being deliberate about every expansion that we did.
– To know that we had work in place to support those people.
– To understand what we were trying to accomplish with every project with were chasing. We didn’t want to say yes to something we didn’t know how to do. You don’t get a lot of second chances in this industry.
We had to be certain that we could deliver. We always wanted to do something better with each project. We tried to research and think clearly. I’m a process oriented person. I want to know how something is going to be done.
The Cat in the Hat was one of the first features we’ve ever done. We had a certain number of effects. We set up our pipeline around those things. I used my knowledge and we tried to forge our own way. Allan is always the one who’s excitable about the projects. He wants to build the company bigger. I was always the one who wanted to keep the company smaller. Between the two of us, we managed to find the middle ground.
[-42:11] Allan: I was talking to a Supervisor at Sony yesterday. We were talking about Frantic Films. I’ve always felt that combination worked really well. One person wants to do everything, while the other wants to taper things back.
Berj: I think that it’s important that everybody does what they’re best at. If decisions come down to financial matters, Allan is the one who knows that very well. He will come to me for an opinion, but he’s the one who makes those decisions. On the technology side, I will go to Allan for his advice. But at the end of the day, I’m the one who has to make it work with my team — and stick by them. I think it’s a good thing to think about when setting up a company. Everybody needs to know what they are good at. It makes for a smooth process.
[-39:41] Allan: You, guys, have worked on so many massive projects. What were some of the more challenging projects in the history of Soho VFX.
Berj: I think that we’ve gotten a lot of big effects that we had to do. And at each step, it was the most challenging thing we’ve ever done. But the first exposure we had to that world was Fantastic Four, the first one. We took charge of Mr. Fantastic character. That was a big, big fully digital character for us who is an important part of the movie. Writing all the software for it and writing all the crazy rigging for it (with Mr. Fantastic’s stretching arms and body) — it was a rigging and animation challenge.
[-37:29] Allan: How did you guys approach it?
Berj: It was kind of crazy combination of a bunch of things that was sort of a Spline IK thing but that would allow us to insert an infinite number of handles. If it needed to wrap around, you’d need maybe a hundred handles. We gave the animators the ability to insert a handle and how much stretching and twisting would happen. His torso would still have to look like a torso. The skin had to stretch in a sensible way. We needed to replicate the texture to make it look like real skin. We’ve used that rig a few times. The software did catch up. But this was in 2004.
[-36:13] Allan: I remember I saw in in 2005.
Berj: It was fun, it was a lot of work. More work than we expected it to be. I know the studio put a lot of faith into us. All eyes were on us. There was a certain amount of pride and responsibility. We pushed really hard and made some missteps along the way. We were still only 3 years out from our beginning and we were still learning. If we had to do that again, it would be a simpler process.
[-34:49] Allan: I think all of those lessons are really valuable to make a bullet-proof pipeline for the next project and the next.
Berj: I’m trying to remember the order of things. The next big comic book movie was X-Men 3. We had a bunch of characters and one of them was Quill. He had a bunch of porcupine needles that came out of his body. We had a cyber scan of the actor’s face and we build all the quills. But we had run into all sorts of technical problems. At that time, it was Maya 6.5, if I’m not mistaken. It was constantly breaking. All the shaders would detach. We went through a lot of pain on that. But we went through it. After that, I swore we would never go through that again. So on the next project — which was The Chronicles of Narnia — we did two things. During the DVD release, they wanted to extend the battle scenes. We had to rebuild all the characters to add 60-70 shots. In a panic, I said we couldn’t do it the way we’ve been doing it before. We built a proper pipeline where we wouldn’t depend on referencing. We took all the painful lessons and learned from them. I started writing software. Each piece of the pipeline was finished exactly by when that department needed to start working.
[-31:07] Allan: Perfect!
Berj: Lighting and shading was the worst part. I had to rewrite that software for the second or third time. That project was the birth of our asset management system which served us very well since 2007. In some incarnation, that asset management philosophy has stayed with us.
[-30:21] Allan: I’ve always found that when studios start to sit down and figure out what the bottlenecks are and what tools would be [needed] by each department in a perfect scenario, it changes their business in a pivotal way. I think it’s awesome for you to have done that! What was so different between the out-of-the-box tools and the ones that you developed?
Berj: Well, at that time, there weren’t any. We’ve come into a world where there is any number of tools to manage a pipeline. But back when I was starting that stuff, there was no product to buy. It was about finding the process. Part of it is: If you’re a hammer, everything is a nail. I’m a software developer, I see a problem — I make a piece of software to fix it. I am pretty sure there was no existing product that existed at that time. If I were to do it again, the problem I see with those tools is that I typically don’t want software that changes our way of doing things. I want to change the software to perform to our way of doing things.
And especially for something like a pipeline: the way we get from modeling and film scans to finished composited images is specific to our pipeline. If the software doesn’t work the way we want, it’s going to introduce inefficiencies and frustrations for the artist. Any time that they have those frustrations, it takes away from their work. I’m a big believer artists should do their artistic work. It would never get to the perfection of that idea. But every decision we make along the way tends to it. We still have our original employee, actually. But staff comes and goes. We’ve built up a mindshare as well. The pipeline is the one constant in all of that. Off-the-shelf tools are too invasive in that way; and I want to keep doing it our way. All of our tools grow and develop. I’m a fan of throwing out the old ones, but reusing all the ideas. Rolling out our own tools helps us run a pipeline the way it should work.
[-25:54] Allan: Cool! While we’re on the subject, what are some other challenging projects that stand out for you?
Berj: Oh, wow! I mean, pretty much every comic book movie we’ve ever done has been in some way a ridiculous challenge. The Incredible Hulk was massive. We were at our largest during that time, staff wise. We had to do two sections in that movie: the plant fight where we first see him as the Hulk and the chase along the rooftop, before the final battle. That was the largest amount of all CG work we’ve done. It was a massive challenge, in terms of the size of the assets and having to match the texture maps from Rhythm and Hues. Amazingly, it went more smoothly than Fantastic Four. We learned that we [couldn’t] get into these massive procedural messes. We had to take the bull by the horn. The thing you have to do is you have to constantly learn your lessons. We’ve always promised ourselves: If we’re suffering on this project, we may suffer on the next one — but it can’t be in the same way.
[-23:33] Allan: Everything with business and production is about the lessons you learn from it. You confidently know you won’t repeat these same mistakes and you’re equipped to handle them.
Berj: We sort of try to make that promise to our staff as well. If they had a tough time on something, we try to make sure they won’t be struggling through the same swamp. We’ll always try to improve things. And I think the artists enjoy working in that environment.
[-22:31] Allan: That’s so great! What were the contributions you’ve made on Logan?
Berj: Logan was a real treasure to work on! We were again working with [Producer] Kurt Williams. We have a long history of working with him. We’re working with him again on X-Men. It was a remarkable project to work on, from beginning to end. Our responsibility was to do the big single sequence, the chase after Pierce arrives at the plant. We took over that section. There was a digital limo, a digital fence, digital guys on motorcycles; the train was all digital. The robotic hand of Pierce was ours throughout the movie. I always like working on things that could exist in the real world if the technology was available. One of the things was that he was repairing the hand constantly, so not all of the screws matched. It was a really fun idea and to make it work in that world. Also, at the end, Logan was fighting with X-24, the twin, and the little kid crushes the truck on top of that. That was all digital. We had a real truck on set, but we ended up going wth a digital model. It was an amazing movie and an amazing team! Everybody brought their a-game to that movie.
[-19:09] Allan: I love that film! I got mad watching it because I realized how great all the other movies could have been!
Berj: I agree! It really made a superhero movie we haven’t seen before. It was more like a Western epic. Everyone in our office bought on, sign on for it. You could tell while we were making it, all hands on deck!
[-18:19] Allan: That’s cool! In terms of your team, I’m curious about your idea on artists bridging to technical [fields]. Do you think it’s beneficial for artists to learn to code or script?
Berj: It certainly can’t hurt. I come from the world when in the 90s everything was done by generalists and they would do everything from beginning to end.
[-17:40] Allan: I still encourage people to learn to take that path. I’ve never seen people in one discipline until I came to the U.S. You need to understand the whole spectrum.
Berj: And I think of the industry, from the scale point: On Jurassic Park, I think there are 60-70 digital shots on that movie. And now, there are hundreds and thousands. Out of necessity, you have to pipeline things now. Even with us, we have a hundred people. And we have different departments, in their lane. But at the same time, even if people can’t do certain things — at least they understand what’s going on before and after them. So when they make their decisions, they’re at least mindful of that. When you’re building a model, you have to think about if it can be rigged. And the rigger needs to understand what the animator is doing. There has to be communication. We encourage that kind of collaboration.
When you think of a more technical side, we do have a team of coders. At the same time, if someone comes and says, “I wrote this things” — it would be silly of me to discourage that, if it makes your life easier. I’m a bit of a control freak this way. We have a lot of artists trying to find new scripts. But I encourage them to talk to me or someone on the team, so that understand their place in the pipeline. We make a software developer out of them by being more engineer minded. If this tool becomes bigger, we will absorb it to turn into something robust. I think it’s a great thing! It lets people be creative.
[-13:51] Allan: In terms of the artists who apply to your company, do you see any common red flags or bad practices? Artists may not be aware of the mistakes they’re making.
Berj: I’m sort of the wrong person to direct that at. I don’t do a lot of the interviewing. I do know what I see when we’re working on a project.
– I think it’s important for people to understand what their abilities are. It’s not a good thing for anybody to claim they know something — but find out they’re not at the same level.
– We strive for physical realism and we have to have our time constraints to do in. Other industries — like tv — don’t have any time at all. It’s good to learn about the compromises of each industry and understand how they work, especially if you’re going from one industry to another. If I were to go to a commercial house, I would be completely lost because it’s a different set of constraints. It’s important for people to understand that and where they are in the world and in the pipeline.
– Understand what kind of work you’re going to get. If you’re a junior, you’re going to get junior level tasks. But any company rewards good work.
– We don’t expect someone to be perfect the first day they start. But we do expect that they ask for help. One of the worst lessons we learned is believe when people say, “We’re fine.” We do multiple reviews and offer to help. Asking for help is a good thing.
[-10:39] Allan: I was just talking about this yesterday. I think you nailed it on the head. The mentality of “Fake it until you make it” is dangerous. You can shoot yourself in the foot if you say you know a certain software. Learn it a little bit and communicate that you’re willing to learn it more. Nothing positive is going to come from everyone expecting your output to be high. And when asking for help, it doesn’t mean you aren’t capable of doing your job.
Berj: This is a young person’s industry and young people have a lot of pride. A lot of these thoughts aren’t true: It’s a sign of weakness, or they’re going to lose their shot. We try to support our staff and find those worlds if someone has some aptitude. All of these things are management issue. If someone isn’t good at modeling hard surfaces — but holy crap, they’re great at character modeling! It’s not like there is a shortage of shots to pass around. We always sit down and figure out who could handle a shot. That’s on us as owners and managers.
[-06:41] Allan: That’s really cool! My last question is more about the review process at your company?
Berj: Nowadays, we’re more review heavy than rounds heavy. When I was supervising more, we’d sit in a screening room. That’s a bit distracting and it takes too much time. On any one show — on Tomb Raider, most recently — we would have a review first thing in the morning and go over that day’s work. On that show, we were doing the same task. Everybody’s tasks would be the same. We would repeat: a review in the morning, in the afternoon, in the evening. On X-Men, where there was a heavy animation / rigging component, each department had their reviews as well. The leads would bring in their work. The animation sup is in charge of reviewing shots until it gets to the level of quality for a supervisor’s review. We try to bring a team together. I’d look at things beforehand, then we bring the team. Everybody gets the notes, in front of everybody else. I think it’s useful for people to hear all the notes. Off they go, after the review.
[-03:05] Allan: I think it’s important to remember it’s a collaborative process.
Berj: They’re a team and they work together. We try not to go crazy with overtime. There will always be deadline pressure on these shows. But we want our team to feel that they have the support. We have to succeed at this together. It’s the best thing any company can do for its employees.
[-01:43] Allan: Where can people go to find out more about Soho VFX?
Berj: I think our website: www.sohovfx.com. Everything is there: We have some reels, job postings; and it’s a great way to send your resume.
[-01:18] Allan: Thanks again for taking the time to chat!
Berj: Oh, my pleasure! Hopefully, you got some interesting information for people.
I want to thank Berj for taking the time to chat. Next Episode, I will be talking to Brad Alexander at HALON. He got his start working directly with George Lucas.
If you’re interested in the FXTD Mentorship, the registration is open right now.
For the free training, please go to www.allanmckay.com/plasma/.
Please review this Episode on iTunes.
I’ll be back next week. Until then — rock on!
* Bio Written By: Soho VFX