Check out Fireball Training at www.allanmckay.com/fireball
Join the FXTD Mentorship at www.allanmckay.com/fireball
Check out www.VFXRates.com
Episode 87 — Interview with Arman Yahin, CEO of Main Road Post
This is Allan McKay. Welcome to Episode 87. I’m speaking with Arman Yahin from Main Road Post. I’m sure you’ve seen their work. They’re doing some amazing stuff, some phenomenal feature work! They’re based in Russia. Speaking with Arman, who is the CEO, it was really interesting to get an insight into the company, their mindset. I personally nerded out a lot, hearing about they’ve implemented Scrum, a team-based workflow that a lot of companies in Silicon Valley use. I’ve got a lot of books on this. Seeing this applied to a visual effects company is really interesting.
Typically, I see visual effects studios throw in more artists or more hours when something needs to be done. Scrum is a unique way of being able to approach it. I found this really fascinating, especially seeing the amazing shot count Main Road Post is able to put out, in a small amount of time and a core team. This is a really great staple of what can be done. I hope Russia is able to get more work.
Let’s dive in!
FIRST THINGS FIRST:
[-1:49:33] I. Before we dive into this Episode, I quickly want to mention that this is the last week to register for the one-year FXTD Mentorship that I will be mentoring.
We’re going to spend an entire year:
– Doing Live Reviews;
– Creating really high-end visual effects content.
It’s been inspiring to watch so many careers be changed by the Course. This is the first time in a year and half that I’ve been able to open doors to registration again. If you want to be a part of it, go to www.allanmckay.com/fireball. Registration closes at midnight on Friday. I’m really excited to meet everyone!
INTERVIEW WITH ARMAN YAHIN
Main Road Post is a visual effects production studio based in Moscow, Russia, since 2006. The studio has produced visual effects for many international films, including Wanted, Stalingrad, The Night Guard, Attraction. It has also worked on the Opening Ceremony for the 2014 Winter Olympics in Sochi.
A year ago, the company implemented Scrum, a team-based workflow that a lot of Silicon Valley companies use. In this Episode, Allan McKay talks with the CEO of Main Road Post Arman Yahin about the company’s history and films; as well as their philosophy on project and team management.
Main Road Post’s Website: http://www.mrpost.ru
Main Road Post Reel: https://www.youtube.com/watch?v=pX4Z-_zCRmk
Main Road Post Stalingrad VFX Reel: https://www.youtube.com/watch?v=BGA73iBDLMA
Main Road Post VFX Breakdown for the 2014 Sochi Winter Olympics: https://vimeo.com/114298918
Main Road Post archives on Art of VFX: http://www.artofvfx.com/tag/main-road-post/
Scrum Alliance’s Website: https://www.scrumalliance.org/why-scrum
[-1:47:54] Allan: First, I’ll say thanks again for doing this. Do you want to give a quick introduction to who you are and who Main Road Post is?
Arman: We are now a 60-people company. It’s relatively big [for] Russia. I know it’s not big in terms of the Hollywood visual effects houses. We started about 10 years ago. We were 6 people [back then]. We work only on movies and tv series; twice we’ve worked on video game cinematics.
It’s a great honor for me to talk with you! Here at our company, we know [your work]. Two years ago, we read your articles about productivity:
- 10 Killer Time-Management Tips for Artists: https://magazine.artstation.com/2015/03/10-killer-time-management-tips-artists/
- Three Great Career Strategies to Learn from Your Boss: https://magazine.artstation.com/2015/05/killer-career-strategies-learn-managers/
We’ve translated it into Russian for our guys.
[-1:46:23] Allan: That’s so awesome! That means a lot! When I wrote those articles, I didn’t realize the impact they would have. A lot of studio owners took them to heart. I think that 60 people is a good, intimate number. Sometimes, 2,000 people is a good thing if you have management to apply good structure. A lot of time, once you go beyond a hundred people, the vibe of the studio changes. I think 60 is a good number to keep.
Arman: I think we will expand. We hope to expand to 100 by the end of the year because we have a lot of work now.
[-1:45:10] Allan: I’ve been watching you, guys, for a long time and it seems amazing to see the growth of your company. Every project that comes out [of your company] breaks a new ground.
Arman: Thank you!
[-1:44:54] Allan: How did you initially get started?
Arman: The core team of Main Road Post started at another company. We were a CG department in a post-production company. There were some difficult times in that company, and the CG department was [dissolved]. We moved under the helm of one new movie company. That was 11 years ago. I think 7 years ago, we moved — to our independence! It was [on] a project called August. Eighth. (http://www.imdb.com/title/tt2094767/combined). The Partner in our company was the director and producer of that project: Dzhanik Fayziev. It was a project that pushed us to grow to 35 people.
[-1:43:21] Allan: That’s really awesome! What’s Russia’s industry like right now: commercial and movie industry?
Arman: There are not many movies with a lot of visual effects. Until last year, we were the [kind of a] company that made one big movie per year (like Stalingrad), and two smaller projects, like a tv series. Last year, there were some changes, I don’t know why. We have a Cinema Fund — the government’s investment in the film business. Our previous work was about war movies. Last year, there were fantasy and sci-fi movies. Maybe producers decided to move [to that genre].
Last year, we worked on four big films. It was a challenge for us. We started to work as a main vendor, not only with our team but with smaller companies. Previous years, we had 60-90 people in the credit block. Last year, we had 200 people in the credits. [The entire] production was 255 people.
[-1:40:34] Allan: That’s really great that you’re able to become the lifeline for other small studios there by outsourcing. I think it’s great to position your company as the big vendor that’s able to bring these big movies.
Arman: I think that too. It’s really hard because a lot of guys who work with us don’t have a lot of experience with the quality needed for these films. There is a lot of painful work.
[-1:39:28] Allan: Yeah, that’s the tricky part. Australia, say, 15 years ago was the same. A lot of us worked in commercials, but there was a new quality in film. Everything needs:
- to be efficiently made;
- to have a certain level of quality.
There is a bigger bar to cross. With any industry that’s maturing, there’s always going to be that hurdle: Being able to manage bigger teams, bigger shots and sequences. There is a lot of growth that happens when these projects land on your lap.
Arman: Yes, yes! You’re right.
[-1:38:30] Allan: While you’re talking about that — having 4 feature films, instead of one per year — how do you handle the growth so quickly?
Arman: It was a couple years ago when the Russian ruble was falling. Because we were buying software and hardware, it was really tricky. It became 2-3 times more expensive. The hardware is from Europe and the budgets became smaller [because of the competition]. If we try to compare with Hollywood, they were tiny budgets.
It was a moment for us to decide: Do we cut our expenses and become a smaller company OR we change and diversify our projects? When you have a big project, you have big money problems because the producers have [higher risks]. We understand that! But for us, it’s a problem. On big projects they have gaps in payments.
So we decided to change:
– We started working with smaller companies;
– And we changed our company [from within].
When we became at 50 people, we had big communication problems within our company, between the departments. At that time, we decided [to implement] Scrum in our company:
– We decided to get rid of departments in our company.
– We decided to move away from being too specialized.
– We decided to have “T-Shape Skills”: [The vertical bar of the “T” represents the depth of skills and expertise, whereas the horizontal bar is the ability to collaborate across disciplines with experts in other areas.]
– We got rid of the names: modeling, shading artist, etc.
It’s been a year since we’ve transformed the company. We’ve split into smaller teams (of 3-7 people).
[-1:32:47] Allan: So how do you build a team if everyone is a generalist? Is it just by sequence of shots?
Arman: Scrum says the team must get organized by itself. We have a few great teams now.
– Each team must be cross-functional.
– The average task is a sequence.
– The team decides who will do which task.
– It’s an interesting experience. We’ve had some problems, someone [even] left our company.
[-1:31:25] Allan: There are so many things I want to comment on! In regards to people leaving was in regards to the change caused by Scrum.
Arman: Yes, it’s very different because in our company, the core software is Houdini. The artists are most of the times generalists and Scrum is really unusual. It’s a good principle for how to work in a team: psychological, social. Some people don’t want to work like that. In Scrum, the responsibility [falls] not only on Supervisors, but on the entire team. It’s not comfortable for everybody.
[-1:29:51] Allan: You’re right! Not everyone wants to be a leader. Not everyone wants to move up. Most people are happy wherever they are in the hierarchy. If you give everyone a chance to think independently, some people don’t want that.
Arman: It’s not only the management because I think we give people the right to decide how to work, how to make cool effects. In Scrum, the team decides how to make it, not just management. It’s hard for some people because it’s based on communication and empathy. Empathy is a huge problem in the visual effects industry because a lot of people just think about what they want.
[-1:28:12] Allan: It’s a two-way street. The articles I’ve put out out years ago were about management utilizing people more effectively using empathy. But at the same time, like you said, artists need to be responsible. Artists managing themselves — in terms of independent thought or having to organize themselves and be reactive — is more about self-management. Some people don’t want that. The fact that you had a few people, it’s natural selection. In a way, that’s a good thing!
Arman: It’s not bad they left. And because everyone in visual effects dreams of working for a bigger company (in London, the U.S., Canada), Russian artists try to build their portfolio. In Scrum, you need to decide what the project needs [which is not effective for someone’s reel].
[-1:25:26] Allan: Going back to some things you’ve said, one thing I’ve found with smaller developing companies. In Australia, everyone was a generalist. Coming here [to the U.S.], everyone was a specialist. From a management point of you, I always feel that in the beginning, you need to start as a generalist before you specialize so that you can have a full understanding of what’s going on around you. The higher you go up, the more you loop back to being a generalist. Say, you’ve evolved to being a CG Supervisor. You need to know everything again. You end up having your hand in everything again. The people who have a full understanding of their job, it gives them more power to do their job a lot better.
Arman: You’re totally right! I think the structure of big companies is not effective. Everything changes a lot. In Scrum, generalists are more flexible. When you work as a specialist in a big company, that company is not flexible. There is a lot of wasting of money for something you don’t need. Some artists are making something, but they can’t see the whole picture; so they’re over-engineering, or over-making, I don’t know.
[-1:22:45] Allan: Yes, fixating on one thing and not seeing how important that one thing [actually is] when we have 200 more shots to get through.
Arman: It’s more of an organic way to work. I can talk about this for hours!
[-1:22:17] Allan: I think it’s amazing that you as a business owner can look at the state of where things are headed and make a big change. Being able to have that growth and be able to adjust — and survive and grow. Having that synergy and teamwork is really important. I think it’s really impressive you had the foresight to change the company for the long term.
Arman: Thank you! I’m not only a businessman. I started as a compositor and I [studied] directing. I worked as an artist for years.
[-1:20:27] Allan: It’s pretty clear that you understand how production works. As a manager, the more you understand what people are doing — the more you get involved. For artists on your team — for people who want to apply to your company — what are the things you look for before you hire [them]?
Arman: It’s funny that you ask. Now, the most important vacancy we have — is Houdini artists. We have a lot of great modelers, texture artists outside of our company [whom we can hire] — and who understand our pipeline. Now we’re trying to hire experienced Houdini artists. It’s not easy in Russia.
We also try to find artists who can work in Scrum:
– They have good communication skills.
– They’re nice people. (It’s really important to be nice because we’re spending a lot of time together at work.)
– They can work within a team. It’s like having smaller studios within a big studio.
[-1:17:38] Allan: I’ve interviewed a lot studios recently [whose artists] talk about how important it is to hire people who aren’t difficult to work with. You can’t work with someone who is focused on themselves. So how can you tell all that, during the hiring process?
Arman: I work with a big number of people so I understand how the guy sitting in front of me will behave. I’m not the only one in the interview, so I ask the leads to be a part of it. We speak with each other:
– Is he nice?
– Is he a good artist?
It’s a hard process. It’s not easy to understand how someone will work and how someone will behave. But it’s an important process! There are no shortcuts. We try to ask the people who work with them, who’ve worked with them.
[-1:15:42] Allan: With Weta, they rely heavily on references. It’s a lot cheaper to pick up the phone and interview someone who’s worked [with an interviewee]. Google, for instance, is constantly testing their hiring process. I think the record right now is 24 interviews once for one job, where they kept interviewing the same person. Obviously, we can’t do that in VFX. Usually, you call them up panicking: “We need that shot done in a week, can you come in?” For me it’s about: Can you have a beer with that person at the end of the day? If it’s 3:00 in the morning, would you be willing to help them?
Talking about Wanted, how did that come to be? Did you have a relationship with Timur Bekmambetov or the VFX Sup?
Arman: It was a really big project for Russia. Timur is very fearless. He decided to move all the VFX work to Russia. His company hired almost every company in Russia and in the Ukraine. We were small at that time. It was the first Hollywood project. We’ve previously worked on cinematics for TimeShift, a video game made by a company called Saber Interactive. The director was David Wheeler. When we started working on Wanted, we had an understanding of big American companies. On the post-production, Stefan Fangmeier came onboard [as a Supervisor]. It was a very important moment. It was very cool to work with him: He’s a star in visual effects. It was great!
We made 100 shots on this film. If you remember the movie: the training on the train. It was four sequences. The sequences were done by us. The project came to us because at that moment we were known by the movie industry in Russia.
[-1:11:09] Allan: At the time, that would’ve put you on a map. Having a Hollywood film is one of those pieces you can put on your chest and say you’ve moved the next level.
Arman: It really was next level because it was different type of quality that they asked. It was great for us. We learned a bunch of principles of how to work.
[-1:10:28] Allan: Just touching base on it for a second, what were the big realizations going from the project you were doing to a feature? How much growth can happen at that time?
Arman: When you’re working in Russia and you’re always watching the big Hollywood blockbusters, for me it was a lot of fantasies. I thought there were some technical tricks, or some super cool software that makes management easier. The main lesson I’ve learned from Stefan was that he works a lot and there is no special software: just email and Photoshop. He worked a lot! The main lesson for me was that there was no magic — just hard work! Iterations, iterations and iterations to achieve great result.
[-1:09:04] Allan: As much as we want to think that there is a magic button with life or work — whether you’re an artist trying to build a career or you’re trying to make a movie — you’re going to have to get your hands dirty and do the hard work.
Arman: That was the main lesson. I don’t know there was some technical stuff: They pushed us to have better keying in compositing or not to touch the main actor’s face, not to change color. On that project, we grew as a specialists because they’ve pushed us to the new level.
[-1:08:01] Allan: So how many people were you at the time of that production?
Arman: Maybe 12 or 15.
[-1:07:49] Allan: Wow! So pretty tight! So when Stalingrad came around, how big of a scale was that film? Did you get the original script or storyboards and go, “Oh, shit! It’s going to be a lot of work!” What were your first thoughts?
Arman: Stalingrad was the project [that was directed] by Fedor Bondarchuk. We worked with him on his previous project called The Inhabited Island. It was a big Russian sci-fi film. We weren’t the main vendor on that film. We’ve made a few sequences. After that project, the producers chose us as the main vendor for Stalingrad. We made all the complex shots on this.
We started with a script, on the early development. It was a long process of development, they had a lot of scripts. It was a long process before the project was green lit. We made some concept art for the first script and waited for a couple years. It was the first film in Russia done in native stereo. Movies shot in stereo have longer shots. For us, it was complex. It meant we needed more quality work on the assets and the simulations. In stereo, when you’re shooting explosions against a green screen, it won’t work.
[-1:04:56] Allan: Because it has no depth.
Arman: We started to work on RnD for fires and explosions, for the huge simulations. It was challenging and interesting. We were happy because it was a big moment for us. In stereo, the main difference is in compositing and the clean-up work. You’re compositing for two eyes.
[-1:03:50] Allan: You need to find the right width.
Arman: Yes, so you must be more accurate.
[-1:03:39] Allan: Were there specific tools you used for stereo?
Arman: No, for compositing we used Nuke. Inside Nuke, we used Ocula.
[-1:03:23] Allan: We went through the same thing with Flight and a few other films. With stereo, we used Ocula.
Arman: At that time, in Russia, there was no company that could convert a big movie.
[-1:02:57] Allan: So you would have to transfer it outside of Russia?
Arman: No, no! That’s why the producers decided not to convert anything but shoot everything in stereo.
[-1:02:35] Allan: What is your opinion about that now? I always thought it was the right idea because you’re getting everything in depth. Talking to a few VFX Sups, a lot of people are starting to lean toward stereo conversions. (I hate stereo, by the way!)
Arman: After Stalingrad, Attraction was converted. I thought the native stereo was better. The conversion looked really comfortable. We provided our conversion company with a lot of masks. For me, Attraction had great stereo. A few days ago, I watched the second part of Guardians of the Galaxy. I loved the movie, but the conversion was really painful for the eyes. It’s really deep and not comfortable.
[-1:00:34] Allan: It’s one thing to critique work. With stereo, it’s not something you can see, it’s something you have to feel.
Arman: I don’t know. Maybe the big depth is hard for the eyes.
[-59:38] Allan: I remember watching Avatar at the Skywalker Ranch. Everyone had a headache the next day.
Arman: But Avatar has positive space.
[-58:41] Allan: Stalingrad was really amazing! Are there any shots or sequences you’re particularly proud of?
Arman: I think we’re proud of everything. There were no easy shots. There were a lot of sequences with fire and explosions; a lot of set extension work. Every shot was really tricky. Making the show reel was a tricky part [in] itself. We made it for 2.5 months, for half of the company. I thought it was the first and last we’d make something like this. But when we put it on the internet, it had a huge success.
[-57:25] Allan: I think it’s so worth putting in the time, to correctly show off what you’ve done. Marketing is so critical to the growth of the company. Even for the CG industry, it’s so important! It looked amazing!
Arman: Thank you! After the Stalingrad show reel was on the internet, we had a lot of emails from a lot of artists from every big company.
[-56:24] Allan: Yeah, what was the reception like?
Arman: It was received very warmly. We thought we’d have a lot of Hollywood work, but that wasn’t [the case]. After Stalingrad, we had a long talk with Chinese producers. That’s what we’re working on [right now].
[-55:42] Allan: It’s interesting! You can do amazing work, but it’s still about relationship building and [it takes] time to break out that bubble. It’s starting to change now. China is a new playing field.
Arman: I talked with a couple of VFX producers in Hollywood. The main reason for that was in tax subsidies. We have those in Russia, for the Russian cinema. There are no subsidies for the outside projects.
[-53:28] Allan: Australia has done it on and off for a few year. As soon as the subsidies stopped, the work dried up. We encourage Australian made films. I think it’s so important to not just pump out Hollywood movies.
Arman: I think that subsidies are distorting the business.
[-52:49] Allan: Let’s talk about Stalingrad for a few more seconds. How many were on the team, at that point?
Arman: We didn’t grow for Stalingrad. It was a long process. The shooting process was split into two parts. We started working on RnD when the principal shooting started. It was a big scene with burning soldiers. We started to work on it. When the second part was shot, there was a long editing process. We had 30 people on our team.
[-51:08] Allan: Wow, that’s impressive! Was that around the time you implemented Scrum?
Arman: No, Scrum started last year. Stalingrad was made 4 years ago. We had a few people, but most Houdini artists are compositors.
[-50:24] Allan: Cool! What about The Night Guard? How long after Stalingrad was that?
Arman: The Night Guard we made last year. In the middle of this project, we started in Scrum. It was really painful. It was a period action movie. We started to work with freelancers. We worked on it for three months.
[-49:32] Allan: I guess for that movie, there were a lot more character driven action. Was there a lot of development time, in building a pipeline around that?
Arman: You know, no. We always used to work to make new things. We change the way we work. For The Night Guard, there was a lot of character animation. August. Eight. had creatures [too]. It wasn’t really new.
[-48:14] Allan: Was there any new technology you implemented, to raise the bar in terms of the quality? Or any new process?
Arman: Maybe not. Before we tried to do some research for the fur, fat, muscle assimilations. We made some tests inside the company. We tried out that research on The Night Guard. The new work was in animation because we were mostly fire and explosion company. It wasn’t really new for us, but it was more complex, a new level of quality. There wasn’t much time for this project.
[-46:32] Allan: So it was a 3-month turnaround?
Arman: Yes, it was 3 months. There was some time for the assets, but not so long.
[-46:25] Allan: That’s crazy though, you realize that, right?
Arman: Yes. When they finished editing, it was only 3 months.
[-46:18] Allan: So how many shots in total, in 3 months?
Arman: Eight hundred shots on the movie.
[-46:14] Allan: So 800 shots in 3 months? And how many artists worked on it?
Arman: [There were 184 people] in the credits.
[-45:59] Allan: That alone is amazing! I really like that. I’m going to use Hollywood as a point of comparison. You’re able to do pretty amazing things. [In big studios], people get really comfortable. You, guys, are able to manage 184 people to create 800 shots in 3 months. That’s insane! To manage that many people in 3 months is alone a massive challenge. I’m curious: What was the experience like managing all those teams in such a short amount of time?
Arman: That was the new part for us! We never worked with a [large] number of people before, or with this many outsource companies. There were a lot of small companies. It’s was pretty challenging.
[-44:31] Allan: That’s pretty impressive! You also did a lot of amazing CG environment in that. What time of process did you have for creating that?
Arman: On The Night Guard, we just modeled the environments in Matte Painting. We made a lot of renders, we provided the outsource companies with those. They composited the shots. We didn’t use a procedural CG engine. We used it on Attraction.
[-43:41] Allan: Talking about Attraction, what were the big challenges before you started on the project? Did you realize how big the project was going to be?
Arman: Yes. Because for us it was a very challenging project. We had an alien character we’ve never made before. It was one of the main characters, so [it had to be] very detailed, with a lot of close-ups. It had to be cool looking and cool animated. It wasn’t a creature in the background. We started working with a creative group of the film: Fedor Bondarchuk with whom we made Stalingrad, and the producers. The producers were young guys [ Alexander Andruschenko and Mikhail Vrubel] with whom we worked on a film called Ghost.
It was a family comedy, it was funny and touching. It was about an airplane constructor who dies in the beginning of the film and becomes a ghost. When Fedor decided to direct Attraction, it was a turning point. He loved the action scenes, so the amount of scenes grew. If you compare it to a Hollywood budget film, it was still very small: only 6.5 million dollars. For the whole movie!
[-39:19] Allan: I’ve worked on a few movies. I did Daybreakers with the Spierig Brothers for 20 million dollars. Even that was a stretch. The directors ended up doing 460 shots themselves because they ran out of money. It’s pretty impressive to pull off a movie of your scale for a third of that!
Arman: When I read the script, I knew we’d be challenged. There wasn’t a lot of time for it. After editing was finished, we had 4.5 months. The whole period of working on this film — script writing, pitching ideas, making the previs, the animatics, how we will shoot something — was about 1.5 year. But the main work was done in 4.5 months.
[-37:45] Allan: When this was shot initially, did you find previs, in a lot of cases, saved a lot of time; how to cut corners?
Arman: Yes, yes, yes! We started with previes. One of the producers Alexander Andruschenko was the editor of the movie. We started to work on action scenes with him and share idea. When we started to prep the shooting, there was an idea to go to a military polygon and dig a big trench. We suggested shooting on a blue screen stage and everything else we would make in CG. It was risky but they went for it. It saved us a lot of money.
[-35:56] Allan: Let’s talk about that for a minute: Did you find that they were more receptive to your coming in and having a say? On a typical Hollywood production, VFX people have to tiptoe around everyone else doing their work. Did you find that knowing a director and being in a different environment, they were more open to your suggestions?
Arman: Yes! There was a lot of trust in us. They worked with us as partners. They understood that visual effects were an important part of the movie. We are very thankful for that! And I love to work with that director: He is a visionary but he likes to delegate. He delegated a lot on us or the editor. We had a lot of trust! It was a very warm and joyful project, but stressful because of time and budget. When we finished the movie, there was a lot of pressure. We didn’t cheat on the deadlines.
[-33:15] Allan: I think that’s pretty amazing when you have that relationship, especially when they realize the importance of whet you’re doing. When you’re working with a 6-million budget, are there any ways you found effective in keeping the budget down?
Arman: We started with Scrum on the previous two projects, but it wasn’t great. You need to start Scrum from the scratch of the project. Attraction we started with Scrum. It wasn’t easy but it was great. We had a couple of small teams. During the hard period of post-production, they worked without supervision. They had their own ideas and found very efficient ones. It was very interesting. The director was open to ideas, not only technical but creative. It was great! Scrum worked great on this project,
[-31:23] Allan: That sounds amazing! I’m just curious, from a technical point: You had an entire CG city. What was the decision to do it entirely in CG, as opposed to live action?
Arman: Because we understood that we needed total freedom and make a lot of destruction, it wasn’t so easy to shoot. We understood that if we shot something, we’d need to make a lot of changes. We came to the producers with the idea of a procedural city. They said okay. I tried to build the city engine in Houdini, one of our artists made the RnD for it. We knew that we couldn’t building because when you’re breaking the building, you need to build everything inside [too]. It was an interesting project. The guys found a traffic algorithm, so we have a lot of cars in the shots, too, animated procedurally. In our showreel, we show the principals. And when we finished our RnD for this, it was fast to build a lot of shots.
[-28:43] Allan: So putting in the work upfront — building a proper pipeline, workflow — allows you to knock things out much quicker, get changes done much quicker, in the long run.
Arman: When they edited the movie, they needed a couple of shots more. We could build those very fast with the environment. We don’t need to shoot it. We made it in four days.
[-27:37] Allan: This is more in general, how big are you on automation, building a proper pipeline, scripting? Is it important for artists to have that knowledge?
Arman: We put a lot of effort into automation. The Partner of our company is Mikhail Lyossin. We’re long time friends for 25 years. He’s a technical genius. He is our pipeline architect. He’s very passionate about planes. We decided to build a full CG pilot. We couldn’t shoot good quality shots when it was flying. Mike did a lot of work on this. No one saw it, but [the pilot’s] badge has Mike’s name.
[-25:37] Allan: I was talking with Jeff Okun the other day, a VFX Sup on a lot of films (www.allanmckay.com/78). He was saying how he sneaks penguins into every movie: Blood Diamond, The Last Samurai, any movie he can. It thought it was cute. In terms of render farm, do you, guys, do cloud rendering? How big is your render farm for projects like this?
Arman: We don’t have a big render farm in our company. On Attraction, we started using cloud rendering because there was a lot of rendering. It’s too expensive to build a new render farm.
[-24:32] Allan: Did you find that it changed how you work at all?
Arman: Yes, now we use cloud rendering.
[-24:27] Allan: What platform did you typically rely on? Were you using Zync or did you have Amazon’s EC3 set up? What cloud services did you find useful?
Arman: That’s a really technical question, but didn’t use Zync. We just build the mirror of our pipeline.
[-23:46] Allan: The advantage of using a cloud, you don’t need to go and build a big farm. Usually, when a production wraps up, if you want more bandwidth, you need to build more space, more computers, [hire] more artists. When the project is done, you’re stuck with this extra bandwidth you don’t need anymore. With the cloud you can turn it on and off. How do you feel at the moment, where the internet is going these days? Working remotely from Russia on a Hollywood movie makes you more capable. You can be anywhere in the world.
Arman: I think it’s easier to work on the internet now. VFX companies are based anywhere. Hollywood movies are more like planet movies.
[-22:16] Allan: I think it’s a great time. You can be anywhere in the world.
Arman: Yes, and the software became standard now. It’s easy to share assets.
[-21:33] Allan: I’ve worked for many large companies and watched them grow. These days, they don’t have the overhead of teams developing new software all the time. Things like Houdini, Nuke are the tools that are capable; wherever they’re not, that’s where you develop new tools. You’re just paying for the licenses.
Arman: There was a time when [my partner] Mike would write a new tool in Houdini. In the next version, they would have it.
[-20:23] Allan: It’s been fascinating to talk about team management and how you’re in Russia and have productions that are shrinking in budget. Software packages are still expensive, but you still managed to grow. That’s so amazing to witness! A lot of companies are focused on getting more jobs in or make artists work more hours. You’ve mentioned having more “T-Shaped” people. That approach is so great. Do you have any other realizations for managing teams?
Arman: Scrum is not just like you start it and you work. You need to always adapt. We worked like that for year and we’re still not happy. When you start to work in Scrum, you need to grow as a team. The team has to build empathy. There will be some stormy times. When we become a good team, they start to build social capital.
Because we’re expanding, there is not easy way to implement it. It’s always a process. The new people whom we hire aren’t familiar with Scrum. For some of them, it’s shocking. You need to act differently within the team. I think if we build a bigger team, at that moment I will be happy. We have great artists who don’t like Scrum. So we need to find a way to adapt. We need to find some kind of a hybrid. In Scrum, you build a team that’s always a team; and our [gypsy] artists want to move around from team to team. We need to find the right balance.
We have consulting companies for Scrum in Moscow. We hired them to help us. We sent some people to training course:
– How to work with conflicts;
– How to work with people [who are resistant] to Scrum.
We’re still working on it.
[-14:16] Allan: Catering to different personalities is always a challenge. You’ve been so forward about making it happen. One last thing: Talking about overtime vs. productivity. When you first read my articles, what were some of the key takeaways appropriate for your team at the time.
Arman: For me, you wrote that article about what I was thinking about at that moment. I’m a workaholic. I was a Senior VFX Sup 14 years ago. My average work time was 14-15 hours a day, seven days a week. When we started our company, I didn’t like overtime. People are burnt out and the morale goes low. We have some overtime now.
[-12:08] Allan: It’s a necessary evil, but it’s more about doing it when it’s necessary. It’s like a sprint. You don’t burn your people out in the beginning. Just doing it for the sake of doing it — you’re stretching them too thin.
Arman: That’s why we started Scrum. When you start in the beginning, you make less efficient choices in the beginning. But closer to deadlines, you make everything more efficient. Scrum has sprints. You have smaller deadlines every week. The main idea was to get rid of overtime work. When you’re working efficiently, you don’t need overtime.
[-10:44] Allan: I wrote that after making ridiculously pointless amount of overtime for three months solid. At one point, you go from being productive to making mistakes and resenting your job. All the mistakes end up costing a lot of time. Having mini goals makes you measure where you’re at. By drip feeding people their responsibilities, you’re able to move forward.
Arman: Yes, and one of the problem is middle management. They’re thinking they understand when the director wants. The overall supervisor makes new comments, then the director makes comments. In Scrum, we show everything to the director. The Supervisor is helping with what we want to build and work directly with the team.
[-08:27] Allan: It’s about making right decisions and not guessing. It’s been so great to talk to you! Seeing how you manage your teams is pretty inspiring. Every studio could learn a lot.
Arman: On Facebook, I have a lot of people who oppose me. I think it is working, but a lot of people don’t understand it.
[-07:03] Allan: People love to oppose anything they don’t understand. There are people who follow the normal path or be innovative. The innovative people are the ones able to grow. I really think you’ve got to think independently. People are always afraid of change. The work you’re doing is proof that it’s working really effectively.
Arman: I’m not thinking that everyone works in a safe way. I see a lot of companies broken and struggling.
[-04:56] Allan: Even when there is resistance to change, you still tried to push forward. It means that you don’t go down that path. There’ve been a lot of psychological studies done on that. Typically, everyone else tries to mimic other people.
For those, who want to find out more about you or Main Road Post, where can they do that?
[-02:09] Allan: Your encouragements [on Scrum] are contagious.
Arman: I’ll be happy if someone writes to us.
[-01:50] Allan: Thank you for taking the time! I know it’s late there.
Arman: Thank you very much! It was a great honor to talk with you!
I hope you enjoyed this Episode. Thank you, Arman, for doing this. I wish the best of luck to Main Road Post! I have no doubt they’ll be killing it and doing some amazing stuff in the upcoming years.
I. If you want to get into the Mentorship, go to www.allanmckay.com/fireball. You’ll be able to:
– Register for the Mentorship;
– Get access to free training!
II. Next Episode, I’ll be doing a Career Intensive. I usually do it in my FXTD Mentorship. This will be the first time I will be doing a public one. Again, sign up at: www.allanmckay.com/fireball.
If you enjoyed this episode, please share it using the social media buttons you see at the bottom of the post.
Also, please leave an honest review for the Allan McKay Podcast on iTunes! Ratings and reviews are extremely helpful and greatly appreciated! They do matter in the rankings of the show, and I read each and every one of them.
And finally, don’t forget to subscribe to the show on iTunes to get automatic updates.