Sean Walker joined the team of Weta Digital in 2005. He has worked on The Hobbit trilogy as well as Iron Man 3, Guardians of the Galaxy Vol. 2, Avengers: Infinity War and Avengers: Endgame.

How did you and Weta Digital get involved on this show?

We got involved very soon after our success on Avengers: Endgame. I am a huge Marvel fan and at the time I had been involved in every Marvel film Weta Digital has worked on. I jumped at the chance to work on Black Widow, being such a long awaited installment in to the MCU, and was grateful to be given the opportunity.

How was the collaboration with director Cate Shortland and Overall VFX Supervisor Geoffrey Baumann?

It was great! Cate is pretty new to the world of VFX but picked it up very quick. Geoff had a huge hand in making the process as seamless as possible for her—watching him work his magic on set was inspiring. I got to spend quite a lot of time on set in London. Geoff and the crew were incredibly welcoming and collaborative the whole way through the shoot. It was one of the highlights of my career.

What were their expectations and approach about the visual effects?

Geoff has a very technical and methodical approach to VFX. We made sure that we started with absolute realism before breaking any « rules ». We based our environment on plates that the team at Marvel had shot in Svalbard, Norway, so the mountains that the gulag is situated beneath exist in real life. We ran animation tests in a virtual helicopter simulator using the same MI8 we have in the film to understand it’s maneuverability. We created avalanche simulations matching speed and volume from reference footage. From there we deviated a little in order to service the story, but we had laid realistic groundwork to underpin everything else we worked on.

What are the sequences made by Weta Digital?

Our work consisted of the Gulag escape, the helicopter crash, Melena’s house and pigs, and the subsequent Taskmaster ambush. It was a nice order of sequences that all happen chronologically in the film, allowing us to keep continuity through all our work.

How did you work with the art department to design the prison and its vast environment?

We started with an amazing amount of concept and art from Charlie and the team at Marvel, much of which we used directly as inspiration for the build of the environment. With so much snow and concrete, we worked with the art department to make sure we were able to add interest and colour where we could without breaking the feel of this being a brutalist icy prison.

What kind of references and influences did you received for this environment?

We received lots artwork for the Gulag environment itself along with lots of brutalist and northern territory photography. Small cities, snow covered towns, and industrial areas that we used for both inspiration and direct reference of buildings and industrial detail. Along with that, we received an enormous amount of footage from their trip to Svalbard which we used to create the extended environment, and the mountain which we generated our avalanche from.

Can you explain in detail about the creation of the prison?

We started with the photogrammetry and the lidar that was captured on set. This consisted of the lower half of the gulag prison yard, and another fragmented set of the bridge that spans over the prison yard. From there we built a digital recreation of the sets as well as an extension for the back wall, and up to the higher levels where the bridge is situated. We replicated all the set details from the exercise equipment to the fine snow that had settled on the whole yard. For the wider environment we used art and reference to create all the surrounding buildings, turrets, fencing and details.

How did you populate and animate the crowd?

For the crowd and the prison riot we spent days in the motion capture studio. We captured enough data to make every shot as unique as possible. Watching each shot multiple times will uncover new mini stories. A friend helping someone up. A prisoner hitting a wall and being disorientated. Little acts that help sell the realism of such a chaotic scene.

Can you elaborates about the shooting of the fights on the walkways?

The walkway fight scene was filmed on a fragmented set elevated 4-5 meters off the ground. This allowed the stunt team some space to work, and low camera angles to be shot as if from the prison yard below. A few fight sequences were prepared by the stunt team and we ran multiple angles for each sequence. We also recreated the exact same sequences using a combination of motion capture and key frame animation so we were able to seamlessly continue the fight sequences while we cut to full CG shots and back to the plate photography.

How did you created the digital doubles?

We created the digital doubles starting with scans captured by Clear Angle on set. We captured each of the main characters as well as an assortment of prisoners and guards. We picked scans that included the widest range of costume elements, so we were able to create dozens of variant prisoners and guards. We were excited to work on Natasha’s new white suit knowing what an iconic costume it was.

A massive avalanche happens during the sequence. Can you explain about its creation and animation?

The avalanche is the largest effect we created for the sequence. It consisted of multiple fracture points from a semi-circular mountain range that collide and combine into a mega avalanche that buries the Gulag and the surrounding environment.

We began with a lot of avalanche footage and testing and figuring out what an avalanche consists of. Aside from a million tons of snow that is. We then broke down the whole effect into multiple parts. The base of which consisted of a broader simulation along with tendrils of ejected chunks of snow. Using Synapse, we then simulated a highly detailed volumetric element from the base, and on top of that, an additional detail pass of fine snow particulate. Doing it in this way, we were able to block out the timing and the animation of the avalanche quickly and adjust to any edit changes that came our way.

Using our renderer Manuka, we were able to handle more data than we have before allowing us to handle larger scale volumetrics. Previously we would have had to separate something like this in to many more passes. We shaded the avalanche and rendered it in a way that allowed us to scatter as much light as possible in the volume. Snow and ice are extremely reflective which gives you that lovely soft but detailed effect. The « icing » on the cake was a beautiful prismatic shading effect to our particulate that gave us just a hint of colour, like when a snow flake is angled at the camera just right.

Weta Digital already did an impressive avalanche for War for the Planet of the Apes. How did you use that experience?

We utilized much of the same technology that we used on War for the Planet of the Apes. The main difference being the upgrades to the technology used to sim and render the avalanche. Manuka can handle much more data now than back on Apes, so we were able to push more detail in to our sims than before. That, along with shorter path tracing times meant we were able to achieve more detail and cleaner renders in a shorter time.

What was the main challenges with the avalanche?

The main challenge was the setup. We’ve worked on many Marvel films, and they’re perfectionists, especially when it comes to the story. I wanted to make sure we were able to facilitate story and edit changes right up until the last minute, so we designed our systems to be flexible and quick, allowing for quick turnaround with timing an animation changes.

The avalanche caused many destructions and explosions. Can you explain in detail about these FX?

When it came to the destruction elements of the avalanche we simulated as much as we could together. Everything needed to interact with everything else. The force of the avalanche would be used to destruct the avalanche, and the subsequent explosions would then affect the avalanche itself. It meant deviating from our more flexible workflow, but it was the best way to achieve the most realistic and integrated effects.

Which shot or sequence was the most challenging?

The most challenging shot was probably the one where the MI8 is pulling out of the avalanche at the last second. Getting a realistic sim in which you could read the downdraft of the rotors inside the avalanche, while telling the story of « did they make it or not », was a tricky balance and a little challenging. The end result turned out amazing I thought, and I was very proud of it.

What is your favorite shot or sequence?

I think my favorite shot is the trucking shot where we travel from left to right with the avalanche hot on the heals of the MI8. It was one of the first that we worked on, included a whole lot of destruction and was beautifully cinematic.

What is your best memory on this show?

I have many! My time on set with the Marvel team was amazing and our production and crew were incredible. It was the smoothest show I’ve worked on.

How long have you worked on this show?

I was on the show for about a total of 12 months

What’s the VFX shots count?

Our final shot count was 461

What was the size of your team?

Just over 300 people.

What is your next project?

It’s been so long that we finished on Black Widow that I’ve actually completed my next project, which was Shang Chi and the Legend of the Ten Rings. My next project I can’t say just yet, but I am keeping the Marvel train rolling.

A big thanks for your time.

// Black Widow – VFX Breakdown – Weta Digital

WANT TO KNOW MORE?
Weta Digital: Dedicated page about Black Widow on Weta Digital website.
Disney+: You can watch now Black Widow on Disney+.

© Vincent Frei – The Art of VFX – 2021

LEAVE A REPLY

Please enter your comment!
Please enter your name here