When this (slightly) Lara Croft-obsessed Editor caught wind of a new collectable being developed to celebrate Tomb Raider’s 25th anniversary, the second most immediate thought after ‘I’ve got to have one of those’ was, ‘Surely, they used 3D printing somewhere?’
Sure enough, in a short video from renowned special effects and prop company Wētā Workshop showing the development of the 1⁄4 scale statue titled ‘Lara Croft The Lost Valley’, Senior 3D Printer Operator David Duke explained how the technology was used to bring this dynamic model of the iconic video game adventurer to life. Speaking to TCT, Duke shared how his role within Wētā Workshop collectibles segment functions as a “translator from digital to physical”, working with in-house sculptors, model making and moulding/casting teams to figure out how a digital 3D asset can be broken down into pieces for manufacturing and assembly, layout the build platforms to get the best print results, and deliver precise representations of the artists' original sculpture. Here, Duke (DD) talks us through the development process, the tools used and where 3D printing makes sense.
TCT: Can you talk about how long Wētā Workshop has been using 3D printing for model making and what machines you’re currently using? Is there any 3D scanning involved?
DD: We have been utilising 3D printing since before I arrived at Wētā Workshop back in 2009 but it's only been an integral part of our pipeline since about 2013. That’s when we got our first reliable in-house machine. Before that we were using a lot more hand-crafted and subtractive manufacturing techniques and only outsourcing parts for 3D printing if it was necessary. But once we got our in-house machine, which was a Stratasys Eden 350V - it changed our manufacturing pipeline completely. Now we run a large range of 3D printers - mostly SLA and FDM. For our Masters Collection editions, we use pretty much exclusively Formlabs printers of which we have Form 2s, Form 3s and Form 3Ls for different purposes. We have scanning available and use it when we need to capture an element off a sculpture or a hero prop. But mostly, if the collectible is being sculpted digitally, we'll be using digital assets that we produce or are given to us by the client.
The 3D model is broken down into suitable pieces for manufacturing and assembly.
TCT: Can you talk about the inspiration behind the Lara Croft The Lost Valley design? How did you move from initial sketches to 3D?
DD: The initial concept sketches came from our senior designer Gus Hunter, who experimented with a variety of poses and compositions. From there, Yorke Yu, one of our sculptors, created rough 3D sketches of the strongest pose options. These were then workshopped by our lead sculptor Gary Hunt, Richard Taylor (Creative Lead and Co-Owner of Wētā Workshop), and the team to work out the best overall composition. From there, Gary started on what would become the final high-resolution collectible, working out finer details like the muscle definition and fabric folds.
Get your FREE print subscription to TCT Magazine.
Exhibit at the UK's definitive and most influential 3D printing and additive manufacturing event, TCT 3Sixty.
TCT: Can you talk us through how 3D printing was used in the development process?
DD: Generally, if we're starting off with digital assets it makes the most sense to continue to sculpt that collectible digitally. And if we're sculpting it digitally, then we know from the beginning that the result will be printed. Both 3D sculpting and hand sculpting have their advantages. And I guess on the 3D side those advantages would be having more control over really fine details and the ability to make quick changes. Many times, characters will have specific texture that is applied to their costume and in a digital pipeline, you can dial those textures up and down as necessary just with a slider. While sculpting is taking place, we usually print texture tests of different parts to make sure the texture is reading properly in the final print. And from an engineering standpoint, working with a 3D model allows us to figure out things like where the armature will sit inside the sculpture, make changes to it quickly if any issues come up, and have it be made at the same time the collectible is being printed. There's also an advantage to being able to see inside the sculpture and having control over how much it will be hollowed out to save on weight - something that is important when you have a character leaping in the air.
Senior 3D Printer Operator David Duke removes a batch of 3D printed prototype parts from a Form 3L.
TCT: It’s a large statue with a lot of detail – were there any challenges in engineering the design to ensure it worked?
DD: The whole piece was a big challenge! Not just because of the scale and amount of detail but also because of the gravity-defying nature of the pose. Suspending one character off another character with minimal points of contact is only possible with some clever engineering. Luckily, we have many minds to figure out these kinds of things. It was a big collaboration across sculpting, printing, model making and moulding/casting to ensure we not only captured the dynamism of the sculpture, but that it was also structural and easy to assemble.
We used lots of big block keys for supporting the characters - basically as big as possible without being noticeable. The piece is 1/4 scale and at the time we didn't have our large format Form 3L printers, so it was a major challenge to cut the model into pieces that were small enough to fit on our printers. In total there ended up being 65 different parts that had to be printed and assembled in model making - with the joins and seam lines for each of those parts needing to be figured out and hidden away.
TCT: How long was the overall development process for this piece?
DD: From concept to final painted prototypes, around 2,800 hours over a 10-month period.
TCT: Can you share how 3D technologies are being leveraged alongside more traditional processes? Where do more traditional methods make sense?
Lara Croft The Lost Valley took around 2,800 hours to develop.
DD: There are a variety of factors which determine whether a collectible is hand-sculpted or created with 3D modelling; time constraints, asset availability, and the artist's schedule are just a few. 3D modelling has a much more involved pipeline with more intermediate steps - for instance, once a collectible is approved a common misconception is that it goes right to the printer. In reality, there is still a lot of work that needs to be done to prepare that model to exist in reality - all the interactions of its parts need to be assessed and subtracted from one another, those parts need to be hollowed and keyed so they will re-assemble correctly once they are off the printer. There is a lot that goes in before you hit print, not to mention the work that goes into cleaning up those prints to a standard acceptable for moulding and casting once they are off the printer. Traditional sculpting bypasses many of those steps - once the sculpture is approved it goes directly to moulding and casting. However, if a last minute scale change is asked for - that's a lot harder to do that in a physical sculpture than in 3D, so there are trade-offs with each.
If we're starting with 3D assets then it makes the most sense in terms of time to incorporate those assets into a 3D model. It also means we have a higher chance of getting talent approval if we have used scan data of the actual person or model.
But a lot of times we'll use an amalgamation of both methods. It's more about using the strengths of each method rather than choosing one over another.
TCT: Given the low volumes of such collectables – 950 in this case – is Wētā Workshop exploring 3D printing for final products or is it primarily a prototyping tool?
DD: We primarily use it as a tool for prototyping. Even at low volumes of production runs, 3D printing isn't really viable yet for output of final product for a variety of reasons. The durability of the 3D printed resins still doesn't match that of castable resins, plus it's much more expensive. But probably the biggest point of difference is the speed of production. It's much faster to create a mould for an object and cast out of it over and over (and a lot more reliable too) than it is to wait hours or sometimes days for just one part to be printed. Until printing gets much faster and more reliable, I don't see it being used to replace moulding/casting for final manufacture.
