One must wonder if Green Lantern is next on Ryan Reynolds hit list for the Merc with a Mouth, considering he resurrected the title character from the cinematic debacle known as X-Men Origins: Wolverine for his latest actioner, Marvel'sDeadpool & Wolverine. Participating in the blood-soaked comic book irreverence is Framestore which was brought onboard by Production VFX Supervisor Swen Gillberg to provide previs, onset support, techvis, postvis and digitally augment 420 shots that ranged from the brutally funny opening sequence to the psychedelic and psychotic third act.When it came to the visualisation team, Swen was excellent in providing video briefs directly from the pitch room rather than relying on written communication, we met often and discussed ideas, explains Kaya Jabar, Senior Visualisation Supervisor, Framestore. He also used our team to iterate on ideas and designs quickly, lens up concepts and present them back to the other HODs [Heads of Departments] to ensure everyone was on the same page. Clear direction and reference were provided for most of the sequences in advance. Having a direct line to Kaya made it even easier to check in to see what avenues had already been explored with Swen and Shawn Levy [director] before adding ideas of our own, states Matt Twyford, VFX Supervisor, Framestore. We were lucky to have had a lot of quality time with Swen on the shoot and having the previs/postvis on hand all the time focused the decision-making. This allowed us to work up the quality of our assets while FPS [Framestore Preproduction Services] fast iterated ideas, hooking into our assets as they developed so the visualisation and visual effects were constantly converging.We relied heavily on motion capture to block out sequences quickly and ensure everything was grounded in reality, remarks Jabar. From a purely technical perspective, managing our scenes on the Oner fight was really difficult with the number of bespoke characters on screen. The uninterrupted flow of the animation proved challenging for real-time playback in viewport, where we tend to live as previs artists. Two unique challenges on the film had to be resolved. Firstly, the Cassandra/Paradox hand intersection and then the Oner, a five thousand frame continuous stunt fight in the City Street set environment, notes Twyford. Cassandras [Emma Corrin] power was shown across multiple closeup and long shots where we deformed Paradoxs [Matthew Macfadyen] whole head and face by pushing Cassandra's hand through it. This sequence required a huge amount of upfront rework for our creature [also human!] pipeline but the payoff as one of the big squirm in your seats moments was great. The Oner was more traditional in its technicalities, but huge in its project management scope. Over 80, five thousand frame takes of motion control, shot exterior through a British winter, generated over a thousand individual artist tasks.Houdini was introduced into postvis to help with timing of the characters and effects. As we are part of the full Framestore pipeline, this was more a workflow rather than tooling challenge, observes Jabar. For the Oner, we designed a way to count the actors on screen at render time and write out a HUD that helped with techvis and planning the actual shoot. One major adjustment for the visual effects workflow and pipeline was for the digital double of Paradox. We knew how to make a fully realistic double with full performance in closeup, but to then shove a hand through that head in a photorealistic and interactive way was not something we normally build for, states Twyford. Although these challenges are usually overcome with time and talent, we knew that the current tools for skin simulation are very much at the back end of the pipeline; this meant we would be showing the skin simulations right at the end of the visual effects process and frighteningly close to the deadlines, especially if any major changes were going to happen. We decided to create a new process where we moved the skin distortions into the animation rigs. This meant that the animators actually animated the skin themselves rather than it being driven by simulations run after the animation. The result was fantastic extra value added by the animators and we were able to show the result right at the front end of the process getting great early feedback from the Filmmakers. The effects artists then focused the post-animation simulations on the fine detail creases and stretches, eyebrow/eyelash and hair interaction. There was no shortage of complex shots to be visualized. For something like the Cold Open we used motion capture of a stunt performer based on initial storyboards, remarks Jabar. We then pitched new ideas on top and fleshed out the cut once stunts had rehearsed the new ideas from the vis. We also helped inform special effects for the carousel section where Deadpool spins, dispatching TVA agents in the centre of a fast circular dolly. The shot with the most iterations is definitely the Oner where we pushed beyond version 200 even in postvis, assembling all the elements and also iterating on the final moment where our heroes jump out of the back of the bus to nail the poses and slow-motion. This shot was also the most intensive in terms of animation and technical visualisation time for previs, as we wanted to really work out every single actor on screen and their action to help guide stunts on the day, while leaving it loose enough to allow them to flesh it out further.The project covered the full gamut of visual effects work from bluescreens, crowd duplication, stunt enhancement, simple and complex environment top ups, photo real simulations, pseudo-science simulations, and creature and digital double work, states Twyford. It was a show that had all the departments busy and challenged.The wide range of visual effects produced for the film also included spaghettification. Spaghettification was introduced originally in Loki Season 2 to show the universe breaking down into strands after major disruptions in the timelines. Only loosely linked to the scientific concept of the same name, it worked well in the Time Ripper sequence as it was established Marvel science and suited the story and visual dynamic. According to Twyford, We had originally developed the look and shots in Loki, so it was great to see it again in another environment as it has such a powerful and dark overtone when you see it develop through a scene. It requires subtle and clever animation by the effects simulation artists and complements the foreground action as it brings darkness and intimacy to the action. The opening title sequence was shot in the UK during summer, which meant that the location had to be practically and digitally dressed with snow. We worked up the dressing to height, added in some falling snow and balanced through the sequence in grade and atmosphere for continuity, reveals Twyford. Wolverines skeleton was a mix of practical and CG. We swapped out various pieces or the entirety of the TVA Minutemen to enhance the violence. Then came the blood. There was no practical blood shot, so everything is a CG simulation custom designed for every impact. The brief was to overdo it at first; some of the initial blood fountains were hilarious but also a bit too ridiculous. The heavy use of bullet time and speed up allowed us to create some beautiful shapes with the blood against snow influenced by Jackson Pollocks work. Technically, this a big challenge as the blood and debris have to interact fully with the environment, the characters and the props. Everything was tightly body tracked, and continuity damage carried through the edits building up to the final full reveal of the carnage. The Time Ripper destruction sequence went through numerous iterations as it evolved. Many of the key elements, like Cassandras destruction, Deadpool and Wolverines damage, the environmental effects both in the upper control room and the chamber below, all were developed dynamically as the edit, performances and visual effects all started coming together, notes Twyford. The only simulation we knew exactly where we were going with was Wolverines costume explosion where the postvis was so awesome we followed their lead. The internal burning for Wolverine was a look developed by the comp team with the cracks and debris coming from effects simulations. Deadpool was partially replaced with our digital double to allow us to blast hot light from inside his body and flare the fibers of his suit. Cassandra died in a million spectacular ways in the ongoing atomization look development of our effects team. We had a fantastic high quality digital double of Emma Corrin including all the skin and subdermal layers through to the skull and the brain. This was blown up, atomized, wind flayed, sandblasted and deconstructed in many unpleasant ways trying to find the right feel for a comedy film where the baddie dies horribly. The final look incorporated an internal plasma overheat with an external sandblasted skin. Thankfully, all over quickly in the final edit.Various Deadpool variants make cameo appearances, including one of a canine persuasion. Dogpool was a digital asset we built to cover any missing performances in the shoot, explains Twyford. Little Peggy and her trainers though were absolute superstars onset, and we got footage for every shot wearing her costume and booties. The original concept did not have her wearing a mask or goggles and we were asked to try and put some doggles on her. This turned out to be massively popular and allowed us to rework the optics to give the googly eye effect; this in turn allowed us to animate her eye performance and open up a new level to her character. In the Deadpool corps sequence her doggles eyes and facial hair are CG over her plate performances. Pinewood Studios in London provided the street environment used in numerous scenes. It included a whole block with the first two stories of buildings, all surrounded by bluescreen, states Twyford. Although it was influenced by the look of New York, it also has a more generic city feel with aspects of Vancouver and Boston. Our role was to extend the set upwards, create a digital city extension for the midground areas and then use a digital matte painting for the distant city horizon. It needed full city life including pedestrians, traffic and believable infrastructure. The set top ups were custom digital models to match Ray Chans [production designer] set design and the city extension used assets previously seen in Marvel productions along with street dressing from our library all laid out to art department references. Cars were a mixture of existing assets and new builds of vehicles to match set cars and buses. All the pedestrians were bluescreen elements shot specifically for the scenes and then dressed into our CG build with our custom Nuke particle tools. Everything was then lit and redressed for each day, twilight and early morning scenes.In the film, blood and gore were not in short supply. All the blood and gore were CG simulations, Twyford notes. With the amount of retimes and big camera moves in the plates we were wary of trying to force in filmed elements, especially when interacting with everything in the scene. Characters, props and the environment lidar were tracked tightly and simulations run pre-retiming to give us a starting reference. Then we creatively tweaked to get the most interesting shapes and framing that worked with the cameras and editorial timing. Once we had a good overall simulation, we worked up the secondary simulations of splashes, soaking into fabric, landing in snow and specific rivulets and drips. The initial brief was to use a comedic amount of blood, so we kept it as clean liquid with no bits or chunks to reduce any unpalatable goriness. A good amount of time was spent fine tuning the surface tension characteristics to allow us a slightly cartoon graphic feel to the flying blood shapes and with a tight shutter angle, every frame looks like it came directly from the pages of a comic. Our biggest challenge was making sure we always kept one eye on who these characters were, in terms of the timing of the animation and the design of all our elements, reflects Jabar. Everyone truly loved what we were trying to achieve, so we wanted to make sure we never lost sight of that despite the volume of work and the complexity. We were on the project for 18 months and I wanted to ensure I kept largely the exact same team throughout and as a supervisor. Ensuring everyone was motivated and rested enough to keep bringing fresh perspectives and their best work was challenging, but so rewarding, and I think that really showed on screen. Paradox proved to be the biggest technical and creative challenge because of Cassandra pushing her hand through his face. The reference was just a couple of the original comic book frames, and our job was to make it photoreal closeup on a performing actor across a dozen long shots, remarks Twyford. What the comic book frame did give us was the level of distortion and key moments, like the finger coming out of his nostril and eye socket. We then had to design the movement and how her hand reacted to the skin, tendons, bones and cartilage of his face in a realistic way. The result is one of the most memorably uncomfortable seat squirming movie moments and I hope everyone in the audience feels Paradoxs pain. Trevor Hogg is a freelance video editor and writer best known for composing in-depth filmmaker and movie profiles for VFX Voice, Animation Magazine, and British Cinematographer.