Personal thoughts/notes from working on Zootopia 2
blog.yiningkarlli.com
blog.yiningkarlli.com
"To give an example of the amazing amount of detail in Zootopia 2: at one point during production, our rendering team noticed some shots that had incredibly detailed snow with tons of tiny glints, so out of curiosity we opened up the shots to see how the artists had shaded the snow, and we found that they had constructed the snow out of zillions upon zillions of individual ice crystals. We were completely blown away; constructing snow this way was an idea that Disney Research had explored shortly after the first Frozen movie was made in 2013, but at the time it was purely a theoretical research idea, and a decade later our artists were just going ahead and actually doing it. The result in the final film looks absolutely amazing, and on top of that, instead of needing a specialized technology solution to make this approach feasible, in the past decade both our renderer and computers in general have gotten so much faster and our artists have improved their workflows so much that a brute-force solution was good enough to achieve this effect without much trouble at all."
https://www.fxguide.com/fxfeatured/the-tech-of-pixar-part-1-...
Related but unlinked Part 2 on other aspects of Finding Dory:
https://www.fxguide.com/fxfeatured/the-tech-of-pixar-part-2-...
I remember find it mind blowing when I learned that in Brave, the artists weren't just using a texture/displacement mapped surface for the clothing and armor. They were using tori primitives for the chain mail, and curve primitives for the clothing. (I.e., the clothing actually woven out of curve primitives for the threads.)
> I was on the RenderMan team at the time and remember thinking it really neat that our system could stand up to that.
> I remember finding it mind blowing when I learned that in Brave, the artists weren't just using a texture/displacement mapped surface for the clothing and armor. They were using tori primitives for the chain mail, and curve primitives for the clothing. (I.e., the clothing was actually woven out of curve primitives for the threads.)
That sounds mind-blowing. Is this documented anywhere?
The things that Don Hertzfeldt did with line drawings and a vacuum cleaner embrace the medium so much more.
But I also get it. It's easy to fall in love with the technology.
[1]: https://en.wikipedia.org/wiki/Disney_Animation%3A_The_Illusi...
For some reason, if I don't think about it, instinctually I would always describe Overwatch (to take a gaming example) and Zootopia as "simple" graphics. My mind recalls big swathes of primary colours in relatively flat yet cheerful lighting, rounded/smoothed shapes, relatively little complex texturing or surface detail.
It's when I pause overwatch that I start realizing 1. how much detail there is, and 2. How quickly and flawlessly it's rendered on relatively slow computers. And then I start truly appreciating the relentless optimizing work to make it "seem simple and fast" :).
Same thing with Zootopia - I've enjoyed the movies (doesn't hurt that I have two young kids), but they would not come to mind if I were asked to name breakthrough or particularly well animated movies. Yet the detail and work is clearly there once you pause and examine :)
I think an art director would describe them as "readable". When there's a lot of detail and quick motion, it's important that the audience can very quickly recognize what they're looking at and what's happen. Otherwise, it just turns into a big jumble of chaos that the viewer can't follow, like in Michael Bay's Transformer movies.
A big part of the art of movie making is telegraphic a sense of rich realism and complexity while still having everything clearly visually parsable. Doing that when cuts and action are fast is quite difficult.
Doing it well affects every level of the production: the colors assigned to characters so they are separated from the background, wardrobe choices to also keep characters distinct, lighting, set design, texture, animation, focus, the way the camera moves. It all works together to produce one coherent readable scene.
[1] https://steamcdn-a.akamaihd.net/apps/valve/2007/NPAR07_Illus...
https://en.wikipedia.org/wiki/The_Peasants_(2023_film)#Produ...
Also a huge fan of that channel. I think he came back recently to do some more episodes. There's a new channel I found that offers similar reflections upon cinema - willbryanfilms - definitely worth checking out!
So much of movie making is like that. No normal person comes out of a theater saying "wow, the color grading on that movie really helped the drive the main themes along, I particularly appreciated the way it was used to amplify the alienation the main character felt at being betrayed by his life-long friend, and the lighting in that scene really sent that point home". That's all film nerd stuff. But it's the lighting, the color grading, the camera shots, all this subtle stuff that the casual consumer will never cite as their reason for liking or disliking the movie that results in the feelings that were experienced.
They aren't necessary. People still connect with the original Snow White, and while it may have been an absolute technical breakthrough masterstroke for the time, by modern standards it is simple. But used well the details we can muster for a modern production can still go into the general tone of the film; compare the two next to each other while looking for this effect and you may be able to "feel" what I'm talking about.
With my comment I was referring to some things that end up being indistinguishable even if insane number of hours were put into it being photorealistic. For example, take a shot where background is heavily blurred. Maybe those assets took a lot to render, compute, used fancy hair simulations and had a lot of details, but they were very far in the distance and camera choice made them indistinguishable from a static background. This is what I am wondering - where is the balance of not doing things that are bound to not be noticed by anyone.
However it's a lossy compression of sorts, and the transform is of a different nature than we're now used to estimating for digital codecs.
Every frame of that movie is an oil painting that could live on a museum wall. Absolutely astounding effort by the crew that made it.
Counterpoint: Avatar. I have never heard someone say they loved that movie because of the amazing story. They like it because it's very, very, very pretty to look at.
I've been debating watching the movies now that there a few of them out, thinking there must be something there. However, I've never heard anyone actually talk about these movies in real life, which is still a concern.
I don't really enjoy the stories all that much, but the experience is fun.
Although, I usually turn off the water while I'm eating my popcorn.
And even with great graphics why are they always so sterile. Where are my dynamic damage models, deformable terrain, procedural content. One tiny thing that bothered me more than it should have was Doom(2016). Great gameplay but the way the monster corpses disappear just upsets me "How can I be knee deep in the dead if there are no dead?" I mean the original doom and quake kept the corpses around. But now 20 years later they can't?.
The new Donkey Kong game has this. You can smash pretty much everything. However, just like with leaning too much on graphics, it felt like they leaned too much on this. The game was very boring and repetitive. I kept waiting for it get better and eventually gave up on it. Being able to smash everything got old by the time I made it out of the tutorial area.
Contrast this to the newer Zelda games. You can climb pretty much anywhere and go pretty much anywhere. It’s the first game I ever played with that much freedom to explore. But instead of subtracting from the gameplay it added to it, as it opened up new (and seemingly infinite) ways to approach and solve a problem. There were also enough varied game loops that felt distinctly different to not feel repetitive.
I'd also argue that it's much like movie editing: you tend to only notice it when it's crappy.
Secondly the bit about the evolution catching the unnamed studio, likely Pixar in production capability as of the first Zootopia certainly shows up on screen.
It somewhat suggests that Disney is correct in having the two studios compete in renderers rather than share one (even as they unify other parts of the process, such as this article mentions moving to Pixar's Presto tool as one of the things that happened in Zootopia 2's production).
When I first started at Disney Animation, at one point I asked Ed Catmull what the rationale was for staffing two separate rendering teams, and he had an interesting answer. His answer was that it turns out that even when Disney Animation was using RenderMan, the high end needs of the studio still required enough rendering developers/TDs that in terms of cost it was essentially no different than staffing a team to build an in-house renderer, and from that perspective he liked the idea of having two separate teams with different focuses/perspectives so that for hard problems the wider company got two attempts at coming up with good solutions instead of one.
To this day the Hyperion and RenderMan teams work pretty closely together and share a lot of learnings/tech/R&D. The focuses are pretty different between the two renderers, and that’s actually been pretty beneficial to both.
The story with Presto is both a bit different and kind of similar. The two studios are now unified in using Presto, but Disney Animation now has an in-house Presto development team that co-develops Presto with Pixar. The two dev teams focus on the needs of their respective studios but move Presto forward together.
I would argue that Disney is definitily a big IT company and relies a lot on that tech.
Perfect storytelling might need flaweless execution to not distract. cirque du soleil for example are also experts in every single aspect relevant to their show/business. Check out the YT video from their sound manager " Inside the Sound of Cirque du Soleil: Drawn to Life" this is so crazy but it explains so much especially how they control the audiance clapping.
Where I work, it's a b2b service company. We've had CIOs get up and say we're a tech company, but when push comes to shove, the IT org always loses to "the business". The business solutions are what are being sold, they really don't care what the tech is under the hood... even if the tech enables every product to exist at this point.
We were also still operating T1s, lotus notes, windows xp in 2014. So I always took it with a grain of salt.
That path of development lead through the 'Moores Law' era to a point where with present use of physics and material science we have run out of going smaller. Maybe there's some new fundamental breakthrough that would unlock something groundbreaking... However it's going to take a lot of hard exploration of fruitless problem space to stumble across such a lead.
In the short term, in parallel, the best thing for everyone with what we do have is to refine further. Find ways of doing things with more environmentally friendly and cost effective processes. That work isn't glamorous or very profitable though, so it'd be better to fund it as a public commons / national benefit work and ensure it's exploited as an international competitive advantage (possibly shared with allied nations for their similar investments).
I wonder how hamstrung Disney is by their chosen animation style (wide-eyed cutesy characters, rounded edges, bright colours). Given the technical prowess on display here, what could they create if they gave their artists free rein to experiment like in Love, Death, and Robots? Or is it more that these constraints provide structure to work inside?
These inventions are represented at Siggraph and are often also research paper and land in the whole ecosystem (in software like hudini).
LDR live from their ideas and stories and experiments with styles. I don't think its relevant for technology it would just be a tech demo.
I only wish it went further! There are a ton of lessons those of us outside films/games could learn from working in that kind of deadline-consttrained innovative landscape. Tell about how you fought against the rendering deadlines and sped up the snowscape frames by 30% to get it in under the wire!
https://disneyanimation.com/publications/
Or, for a continuously updated list of publications specifically about Disney’s Hyperion Renderer:
Specifically, I see incredible effort and time getting tiny details right on the technical side, but the storyline and dialog seems to not have had the same level of effort applied to it.