VFX Supervisor Wojciech Zielinski talks about adding depth, breadth, and grit to the 2,596 visual effects shots produced for the second season of the hit Paramount+ sci-fi series, adapted from one of the most popular video game franchises of all time.
Worldbuilding, as they say, isn’t for the faint of heart. Just ask VFX Supervisor Wojciech Zielinski, who led the team that created the visual effects for the second season of Halo, the well-loved Paramount+ sci-fi series based on the well-loved Halo video game franchise.
As most readers likely already know, in the series, Master Chief John-117 (Pablo Schreiber) leads his team of elite Spartans against the alien threat known as the Covenant. As humanity’s best hope for winning the war, John-117 discovers his deep connection to a mysterious alien structure that holds the key to humankind’s salvation, or its destruction — the Halo.
In the latest season, two new major characters join the saga, “formidable intelligence operative” James Ackerson (Joseph Morgan) and Talia Perez (Cristina Rodlo), a corporal specializing in linguistics. There are also new environments and a more emotional, grittier approach to the narrative. All of which adds up to an enhanced experience for viewers, and a potentially daunting series of challenges for those tasked with bringing all of this to believable life.
All told, 2,596 VFX shots were produced for Season 2. Vendors on the show included Luma; El Ranchito; MPC; Image Engine; The Yard; Cubica; Monkey Rave Productions; WEFX; Rocket Science; Crafty Apes; IonArt; Torpedo Pictures; Incessant Rain; and FillScrn. Phew! Previs was handled by Pixomondo, The Third Floor, and Perception, who also did concept art. MPC Art also provided concept are work.
And the stalwart responsible for wrangling all those VFX studios’ effort, the above-mentioned Zielinski, virtually sat down with AWN to talk about how it all came to pass.
Dan Sarto: I enjoyed Season 1 of the show, and was waiting very patiently for Season 2, like everyone else. And I think Season 2 was even stronger, especially from a narrative standpoint. What, if anything, was the biggest visual directive coming into Season 2?
Wojciech Zielinski: The most important thing to remember about Season 2 is that there is a totally different approach to the story and there is a tonal change. The season is more centered around the characters, and it has a much more grounded and gritty look. And that reflects the marching orders I received from our showrunner, David Wiener. That was something that we discussed in depth during my first meeting with him, when I was auditioning for the show. He said they wanted to walk away from the look that was established in Season 1. They wanted to actually elevate the show by adding more depth to the characters, and that way hopefully build a larger audience for the show. And what it meant for us was that, as we were inheriting a lot of assets – and I'm talking right now from the visual point of view from Season 1 – we needed to adjust the look or the properties of those assets to fit the new direction.
DS: That said, how many of the assets were you able to carry forward? I think anyone in your position would have to redo some rigs or other stuff just because, even in a few short years, there's new technology or new pipelines. I believe there were also some new visual effects houses that worked on Season 2. How did all of that impact your creation process?
WZ: Let me start by talking about the look of the assets. You're absolutely right that each vendor has a different pipeline. As far as the creatures go, every vendor has a different rigging system, etc. It goes without saying that the vendors who are awarded the job will take the existing assets, and they will ingest them and rework the technical aspects. But what was most important was what's on the surface, what do they look like. And this is what I meant earlier about implementing this modification to make them work with this new, deeper look.
If you compare some of the creatures – for example, the Sangheili warriors and the Jackals – you will notice a few modifications in terms of the texturing and the surface material properties. We wanted to make sure that they felt and looked more believable and more organic. Of course, we also introduced the new characters, Arbiter Var 'Gatanai and Uto 'Mdama. These were the sort of major players that had to be designed from the ground up by the VFX team, in collaboration with the showrunner.
As far as the spaceships and vehicles go, of course we modified them too, to make sure that they were going to serve the story requirements. We also have the rich city asset, which was created in Season 1, that had to be modified in Season 2. In this case, we had not only to elevate the look of the city, but we actually had to basically rebuild or rerig the city for destruction, which happens in Episode 204.
DS: How much design – either concepts or full designs, including any previs – were you provided with, and how much of this new design, or redesign, was largely done at the visual effects production level, working with the showrunner?
WZ: Predominantly, for all the environments, for the world creation in the show, we were provided with concepts done by the production designer. He was in charge, that was his vision. Of course, once it goes into post-production, we may have to modify it slightly to make sure that it’s going to work. As far as the spaceships and the creatures, the designs were totally done by the VFX team.
DS: That's not something that's always done. It would seem like on a show like this, where the visuals are everything in a certain sense, it would be a lot of fun for the VFX folks to be involved at the design level.
WZ: Yeah. That's what keeps us going. That's the most exciting part of it.
DS: As far as the live-action shooting, how much were you able to use, or just extend to produce the visual effects, and how much did you have to redo completely in CG? For example, it looks like there were some live-action plates of mountains and things like that.
WZ: In general, of course, I'm always trying to get as much in-camera as possible, whether it's environment or creatures or some action moments. But we are there for a reason, and we want to make sure that, when we conceive the sequences, our VFX team gets the right parts or right components to allow us to work with that in post. Let's use the final battle between the Arbiter and Master Chief as an example. This was quite a complex scene, from designing the choreography of the battle, to pre-planning the shoot on location, to the post work. Each battle with the creatures were pre-designed, or the choreography was pre-designed, by the director and the stunt coordinator. We shot that particular sequence on locations with the stunt and actors in-situ.
And then we had to basically reproduce that choreography in CG with the CG creature and with the digi-double. Also, we had to augment the environment. As I said, it was quite complex. And for this particular environment, we were shooting at the beginning of spring in Italy, on a border with Slovenia. We shot that in spring and we didn't get the super-lush greenery, which you would expect from the Halo world. I mean, this is something which gamers would bump on right away. A couple of months after we shot the show, the VFX team went to that location again, and basically we shot plates and tiles, which we used to augment the action plates.
DS: Apart from this final battle, what are some other visual effects that you think really represent your work on the show? There's a lot of creature work, a lot of creatures on screen with humans up close. Was that a big challenge?
WZ: The same principles that apply to the final duel between Arbiter and Master Chief on Halo apply to every action sequence with the creatures. It is all about the planning, the design, and then the execution. And what's also important for us for the VFX is the data collection, by which I mean the motion capture. And knowing how intricate these battles are, we decided to try a new approach. Traditionally, you shoot those action sequences, and then you take the stunt team to a mock-up stage, and then basically you try to reproduce it. But, with today's technology, we thought we could be smarter. I'm not sure if this is something new, but we figured, why don't we make the actors and the stunt performers wear the mocap suits? Let's capture the mock-up as we are actually shooting the shots. That way we are going to get instant feedback from the data.
And the data is going to be as precise as possible because it's going to match exactly the action that we have in plate. Of course, later on, we used a small mock-up station after the shoot to capture additional performances; but 95% of that data actually came from the shoot on a particular set.
DS: The Covenant creatures are quite big. How did you handle them in the live shoot? Did you have someone in a MoCap suit?
WZ: Basically we had two different scenarios. As you know from the show, our creatures are larger than humans. In static moments, to compensate for the height, we had our actors and stunt performers wear stilts in order to get the right relationship. In the very dynamic actions, obviously the stunt performers couldn't wear stilts for safety reasons. We came up with a head rig, which was basically a bunch of wires attached to the back of the stunt performers, with a tennis ball at the top to give the actor a perfect reference for the eye line. Also, every time that we had a direct interaction, like a body contact between the Spartans and the creatures, we had to use a full CG approach and rebuild the moment with digi-doubles.
DS: With digital hero characters – especially bigger characters – that are onscreen for a considerable amount of time, as well as onscreen with human characters, the issue of weight is really important for believability. How did you ensure that these characters would convey the requisite weight to go with their size?
WZ: As I said earlier, as a foundation for our work, we were using motion-capture data, but of course humans can only do so much. So it's all about finding the talented animators who infuse those creatures with life. I mean, it's super-challenging. Of course, we go through multiple iterations. It's a very organic process, and we discover things in the process. But it all goes down to the talent of the animators.
DS: Is there anything else you want to highlight with regard to the work your teams did?
WZ: We had quite a complex build in Episode 201. This is the sanctuary, when we see the Covenant and the Arbiter for the first time. This world is a hybrid of plate work and full CG environments. We shot that in Iceland and on a bluescreen stage in Hungary. It required a lot of that data collection for the environment as we were shooting in Iceland. We had a dedicated team that was literally scanning the rocks, scanning the locations on which we were working. There was a lot of plate photography done for different moments in the sequences. In the end, I hope that we achieved our goal – that we basically created a consistent and seamless environment from the start of the sequence to the end of the sequence.
Another very challenging thing in the sequence was the glassing, because we didn't see it in Season 1. And we had a team of talented artists working relentlessly on the look of the glassing. We started using the game reference as a foundation. But then we had to develop that on a concept level before we actually started shooting, because the final look would let us know how we were going to execute the shots, what support we needed from the special effects team, what support we needed from our director of photography, in terms of interactive lighting. There were many, many elements, many components that we had to coordinate to make this look as exciting as possible.
Another big environment is in Episode 204, which ends in the destruction of the city. We see it at the beginning of the episode – this is the moment when the Covenant forces begin to attack the city, and one of the first big crane shots revealing the destruction is a fully CG shot. Then, later on, we follow the characters to different parts of the city as they battle the creatures, which is again a hybrid of practical locations and CG environments. Then, later on, during the bridge battle, we used a blue-screen stage, where we worked with just a small partial set of a bridge, where the railings and everything around them was blue. And we basically had to create the world and the destruction of the world around it.
