From 2D Supervisor on 2018’s Venom to VFX Supervisor on The Last Dance, learn more about David Lee’s creative journey working on Sony’s hit anti-hero trilogy!
Just over a month ago, the final instalment of the Venom trilogy hit theatres – and was, appropriately, the top movie worldwide for 3 weeks following its release!
As lead VFX partner on the previous two films in the trilogy, DNEG was honoured to return once again for the final instalment. Our work was led by VFX Supervisor David Lee, who previously worked as 2D Supervisor on the first film! From creative challenges to new tools, we sat down with David to hear more about his journey with the franchise. Read it here:
Hi Dave and thank you for joining us! How did it feel having your role evolve as it did throughout the films?
There is always a sense of trepidation when moving roles, a healthy respect of the unknown which I think is something to be embraced. There’s quite a leap between watching people make decisions in roles above you, and then being the one ultimately responsible for similar decisions when you get that opportunity. Generally, however, there is a real excitement to keep facing new challenges and learn new skills. And helpfully, I’ve always been lucky to have the support of a great group of people around me at every level throughout my career!
Can you share your journey in the visual effects industry, and how you transitioned from a 2D Supervisor to a VFX Supervisor?
I began in visual effects back in 2005 in New Zealand. Having graduated from Film School, I began filming my own short films before learning After Effects online, and moving into animated shorts. This ultimately led to me getting a place at Weta as a Compositor, helped by good timing and good fortune. From there, I freelanced around New Zealand for a few years, then moved over to the UK and worked on commercials before the urge to move into longer forms led me to Cinesite, where John Carter was in full swing. After a few years there, I moved over to DNEG in a Lead Comp role on Man of Steel, and eventually into 2D Supervisor on the original Venom, then DFX Supervisor on Tenet, and VFX Supervisor on Meg 2: The Trench and Venom: The Last Dance.
What lessons did you learn from working on the first Venom that you applied to the final instalment?
I think generally it was a more iterative process from the first films. Via the wonders of hindsight, we could make adjustments to the set ups that had challenged us last time!
For example, during Look Development for Venom, we created a variant of him for shots where he is small in screenspace, which we gave a different specular response, as it was previously harder to achieve the classic Venom look in these examples. And really, it’s many smaller scale adjustments that holistically make a difference. We would know which attributes might be more prone to client notes, so we could ensure they were particularly robust while we had the time early on to do so.
Have you seen any significant changes in visual effects technology or processes between the first and final films of the trilogy?
The first Venom was released in 2018, so we’re looking at a time span of about 6 years. In that time, the two things that jump out are machine learning, and automation.
Automation certainly is not a new process, but development of this within DNEG has come along dramatically in its scope and ease. Classically, this has always been done in a more manual sense. For example, in comp, a lead may create a template that can then be copied and passed around to artists to give a solid base for shot-specific work. In the last few years however, this type of workflow has been integrated much more deeply into the pipeline, with more advanced tools to allow individual artists to work on many more shots with a much higher level of consistency than has been achieved in the past. These artists can also maintain ownership of these shots for a much longer duration! It’s a much more efficient and consistent workflow. It also is a boon for late-in-the-day sequence changes, as in an ideal situation, we only have to make that change once, and it will flow through a multitude of shots without derailing the schedule.
Machine learning is coming along at a tremendous pace. This has moved from the realms of reading about it, to actively starting to use it. We are still at the very beginning of this technology, but even the use of the copycat tool in Nuke to bash out rough holdout mattes is something that didn’t exist during the original film. We do have some way to go with these new tools to get the level of quality and flexibility we need for final feature use, but the time between iterations is drastically reducing – and there will only be more of this in the coming years!
What are some of the creative or technical challenges around working on a franchise like Venom, and how did you and your team overcome them?
I think the main challenge of franchise films like Venom is making sure you are ready for anything. There are a number of different creative inputs in shows like these, so preparation is key to ensure that any curve balls that come down the line are able to be incorporated into the show with as little stress as possible for everyone involved. So, technology and tooling play a large part with this, and we placed additional emphasis on every department to ensure that the fundamental set ups were solid, and flexibility was built into every approach. Now obviously this isn’t always possible, but it is a good rule to live by. Essentially, expect the unexpected.
What advice would you give to aspiring VFX artists looking to work on big projects like this?
I think anything you have passion for, you will put in time with, and naturally become more practised and skilled at. Keep referencing the world around you and use it for inspiration. It holds the answers to everything!
A big thank you to David for his insights! Watch this space and follow us on social media for more Venom news.
Craving more Venom? Enjoy this reel of (just some!) our favourite DNEG shots from each film!