Not only was it made with Blender, the final renders were done with Blenders semi-realtime Eevee engine rather than its max-fidelity Cycles engine. That reduced the compute required by orders of magnitude - the director said a render farm wasn't necessary because his local workstation could produce final-quality 4K frames in 0.5-10 seconds.
"Proper" production renderers like Cycles do look better of course, but having an alternative which is viable on a shoestring budget is very valuable.
Now I'm curious how the film would look if it was rendered in Cycles, there are some lighting aspects that really feel "off". Perhaps now that the film is acclaimed they could release a remaster done in Cycles.
First off, switching to Cycles is probably quite a bit of work. While the renderers are supposed to be interchangeable, since AFAIK Cycles supports more features than Evee, options that previously did not matter with Evee rendering now have to be set for Cycles.
Also, having seen the film, I found the "unrealistic", cartoonish look very much to be a creative choice. Evee can produce much more "realistic" renders than what you see in the movie, but this requires also much more investment into things like assets and textures, otherwise you quickly land in the uncanny valley. So I think switching to Cycles probably would not matter much, unless the creators would also change their creative choices, which would result in a different movie, but not necessarily a better one.
I kind of hope they don't. I like the humble, democratic, FOSS spirit - it's like Dogme 95 / "Vows of Chastity".
"rules to create films based on the traditional values of story, acting, and theme, while excluding the use of elaborate special effects or technology. It was supposedly created as an attempt to "take back power for the directors as artists" as opposed to the movie studio."
I had a negative initial reaction to the animation style but it hooked me in and blew me away. It had virtues far more vital than render quality. In contrast, I bailed on "Inside Out 2" and have no interest retrying. I hope more people are encouraged to create lofi meaningful movies instead of thinking it's the preserve of billion dollar studios and sweat-shop animation factories.
Like the first time I played Super Mario Bros. on an LED screen. Finally I could see each pixel clearly, exactly the way the original artist didn't intend!
Edit: in all seriousness, this makes me wonder: has anyone ever re-orchestrated Beethoven's Fifth? Say, in the orchestration style of Ravel or Strauss? Someone must have done this, even as a joke, and I'd love to hear it. (I know about the "Fifth of Beethoven" disco tune which is great, but that's not what I'm asking about.)
Haven't watched yet but that was my gut reaction. When the engine first got stable released I tried it and was impressed at how quickly it got to a 90% solution, but the now complicated lighting scenarios that it couldn't handle took me back to cycles.
Cycle's renders are beautiful, but 10 minutes per frame can be a hard sell... I wonder, if anybody tried rendering in cycles to output eevee's primitves. I remember that was one of the tricks that architecture rendering community used - just paint with lights in places that a full blown global rendering/path tracing would do.
IIRC, the Blender Foundation's Open Source movies have been rendered on render farms from the very first one, produced over 20 years ago. This predates Cycles/Eevee, but I don't think it's something they'd regress on.
a stupid-simple approach would be to split up the render betweeen machines by manually starting it on each one and setting different frame ranges to render
You're probably referring to Cycles X [1], which if I'm not mistaken has already been released.
It will never be on-par with Eevee's performance though as they are fundamentally different approaches to rendering: Cycles is a physically-based path-tracing engine, while Eevee uses rasterization through OpenGL.
I just had a look at the trailer, and I'm trying not to poo on it's parade, but this thing looks... disappointing - worse than most in-game cut-scenes these days. It doesn't even feel "Artistic", and I'm definitely not a snob for "hyper realistic" types of looks.
The distant and "landscape" views look very nice, and in stark contrast to the game-like and amateur rendering of close up scenes with the animals. They don't even have anti-aliasing and the things look "blocky".
I hope this thing won because of the story and characters, and not its visuals.
There are incredible visuals in the movie, but not because of their realistic details. They are instead incredibly evocative of a mysterious depth behind the relatively small story being told in the movie.
The movie doesn’t look real, but it also doesn’t act real either.
With the amount of utter trash that modern Hollywood puts out, combined with the Oscars always feeling like a "pat on the back for rich snobs," I am just genuinely happy to see something like this win anything at all.
I watched Flow in the theater. On the big screen, the less-polished look definitely comes across as an artistic choice, enhancing the otherworldly quality of the world and scenes it depicts. Combined with the great animation of the animals and the audio, you leave the theater in wonder. I suspect the visual roughness becomes more jarring in a less immersive environment.
I agree, but mainly for some other reasons I won't be listing just now. As much as I find it all very cute and I'm a sucker for this kind of shit (I can already sense I would cry my heart out watching this at some point or another--I know, I'm very sensitive), it kinda looks like a long-winded tech demo or video placeholder.
I did not find Flow to be a technically impressive movie. The animation was very imperfect. The rendering (especially shadows and textures) were off. The whole movie looked like a video game cut scene.
But oh boy, what an amazing cutscene to watch. I'm worried that the story the media is putting forward is that this was an innovative and cutting edge movie - based only on a superficial appreciation of the (stunning) art design. But the real story is how the director worked within his limitations to make something equally enjoyable and meaningful as the other guys.
Most importantly, this movie passed the Actual Kid (TM) test. My 7 year old and his friends sat raptured through the entire movie without any slapstick, pop music numbers, or even dialogue! Not once, but 4 times now!
Yep, grew up on cartoons like Transformers, GI Joe, Thundercats, etc. Looking at them now, they are laughably bad in most respects, but they sparked our imaginations and didn't need to be sophisticated to do it.
A lot of the cartoons I watched as a kid had excellent animation, and a lot was very primitive 3D rendering that looks horrible in comparison. As a kid, I didn't even notice!
I feel like it kind of fits in the same category as Hundreds of Beavers (also a fantastic film), as something using the roughness of low-cost methods as a genuine part of the artistic style.
Recently I reminisced about Blender foundations first(?) effort, Tears of Steel, with the script like "Look, Celia, we have to follow our passions; you have your robotics and I just want to be awesome in space!" - "Why don’t you just admit that you’re freaked out by my robot hand?!"
That's kind of surprising. Academy members are not required to watch all the nominees for Best Animated Feature before voting. In fact they are not require to watch any of them.
Several years ago I remember that after a year where the movie that won best animated was not the one that those in the animation industry overwhelming thought was sure to win some animation industry magazine survived Academy members asking which movie they voted for and why.
What they found was that a large number of the voters thought of animated movies as just for little kids and hadn't actually watched any of the nominees. They picked their vote by whatever they remembered children in their lives watching.
E.g., if they were parents of young children, they'd vote for whatever movie that their kids kept watching over and over. If they no longer had children at home they would ask grandkids or nieces or nephews "what cartoon did you like last year?" and vote for that.
Another factor was that a lot of these people would vote for the one they had heard the most about.
That gives Disney a big advantage. How the heck did Flow overcome that?
Inside Out 2 had a much wider theatrical release in the US, was widely advertised, made $650 million domestic, is the second highest grossing animated movie of all time so far worldwide, and streams on Disney+.
All that should contribute to making it likely that those large numbers of "vote even though they don't watch animated movies" Academy members would have heard of it.
Flow had a small US theatrical release at the end of the year. I didn't see any advertising for it. I'd expect a lot of Academy members hadn't heard of it.
As a guess, maybe Moana 2 is the movie that the kids are repeat streaming. That was not a nominee so maybe those "vote for what my kid watched" voters didn't vote this year and so we actually got a year where quality non-Disney movies had a chance?
Interesting. I loved Flow and I'm glad the stars aligned for it on this particular occasion. This article [1] lists a bunch of other Oscar-related firsts:
* Gints Zilbalodis, who is 30 years old, is the youngest director to win the Oscar for best animated feature.
* Flow is the first fully-European produced and funded film to win the feture animation Oscar.
* Flow is the first dialogue-less film to win the feature animation Oscar.
* Flow, made for under $4 million, is by far the lowest-budget film to ever win the category.
It also says the winner of the animated short category, In the Shadow of the Cypress, was unexpected since the Iranian filmmakers couldn't do any of the usual in-person campaigning of Academy voters due to visa problems.
1. The academy has had a significant increase of young voters in the past 10 years or so. Generally speaking, young voters are more likely to take animation as a "serious" medium.
2. These interviews were always somewhat overstated. Of course some voters have stupid rationales, but I don't think this dominates the academy.
3. Disney's Inside Out 2 was nowhere close to winning the award this year - Flow's biggest competition was The Wild Robot, which did gross far more than Inside Out 2, but far below Inside Out 2.
If you look at the past couple years, The Boy and the Heron (Studio Ghibli) won over Across the Spider-Verse (with Pixar's movie Elemental nowhere close) in 2023, Guillermo del Toro's Pinocchio won over Across the Spider-Verse (with Pixar's movie Turning Red nowhere close) in 2022, etc.
I'm curious what year you're thinking about above. Perhaps Toy Story 4 over Klaus in 2019?
4. The results can still be valid if there’s a lot of random noise in the sample. There are about 10,000 voters here. If 9,000 vote at random and 1,000 watch the films and vote on merit, there’s about a 2% chance of getting a different result than if all 10,000 watched and voted on merit.
I would guess this is, to some degree, a generational shift. The Animated category has only existed for ~30 years and was born from the resentment many in the academy felt toward Beauty and the Beast being nominated alongside supposedly serious films for Best Picture. Each generation following that one has grown up with a more diverse slate of animated films available.
The Oscars are the slowest possible reflection of social change, and I’m sure the perspective you share is still held my many members, but this win holds out some hope for sure.
The Grammy awards for music are the same thing. Members aren't required to listen to the nominated albums, and every member gets to vote in every category.
I had a friend who was a Recording Academy member as a classical musician. He thought it was strange that they asked him to vote for the best hip-hop album since he doesn't listen to hip-hop at all.
So for many of the categories that are a little more niche, it basically turns into a popularity contest, rather than the opinion of true experts.
I adored Flow. It's hard to say it was truly "better" than Inside Out 2. I think part of the calculation has to be that everyone expected Pixar to deliver something top notch so it only really met expectations. Flow was made by a no-name team from Latvia and was really something unique and interesting. I went into it kinda blind with no expectations and was blown away.
I didn't think inside out 2 was a very good movie.
It had good ideas but didn't do very well with them (contrary to the first movie, which was great). I'm not surprised a movie which wasn't "just a sequel" managed to beat Moana and IO2.
They aren't required to watch them, but the voters do all get screeners. They at least have the opportunity to watch it, regardless of whether they've seen it in the theater. They don't vote just for what they've heard of.
The Academy has a reputation for seeking "artistic merit" even at a cost of good entertainment. They're hoping to advance something that didn't do well at the box office. Sometimes that means giving awards to films that turn out to be dogs, but sometimes they manage to promote things that deserve attention.
A lot of Oscar-bait gets a small release at the end of the year, to qualify it for the Oscars. If it gets a nomination, they'll use that as part of a wider campaign later. That's why they send out screeners: they know that many members won't have had a chance to see it in the theater.
For me, Flow's greatest strength was the complete lack of voiceovers, almost like a silent film. No overbearing narrator coercing you. Flow allows you to feel on your own terms without interference.
With most media since the dawn of Hollywood, the internet and now AI, we are accustomed to being told exactly what is happening. Think about how 'laugh tracks' tell you to laugh. The search for an answer or meaning of something is largely taken away from you. Without that instruction you are left to make your own interpretation of things, no delivery of a specific message or theme. This means the movie is experienced differently by everyone. That why it's so great.
As someone who started using Blender before 1.8, posting on the old blender.nl forums before its move to BA, it's just been pretty insane to watch it reach this point. Back then it didn't even have ray tracing, and all of the attempts to make long form videos with it were very very rudimentary.
It shows, Blender has come a long way, but FLOW doesn't look technically incredible. On the otherhand, I just rewatched Shrek recently, and complex graphics isn't everything.
What do people mean with technically impressive? There are Blender renders that look quite incredible though and you just cannot differentiate it from real picture anymore
There are probably some flaws here as well, but you need to study the picture in detail. And Flow used the fast renderer of Blender, not the quality one.
Still, it does have a unique style that is much more interesting than many other animated movies. So what is technically impressive, just throwing more compute at it to make it photorealistic?
I think art style will have a larger impact. In a way it is technically impressive as it didn't need a lot of compute power.
I think people are specifically referring to the movie not looking technically impressive, not that Blender isn't capable of technically impressive renders at all.
No kidding, Shrek probably had to do with 100x less computing (per hour) than a modern production. First Toy Story probably something like 1000x less computing
Great example of how accumulated technical innovations unlock unexpected opportunities. Flow, Shrek and Toy Story have roughly similar technical quality but vastly different price tags. That cost reduction allows more experimentation, which delivers more compelling outcomes.
Yeah, the real time stuff is pretty solid in blender but for very high image quality / realistic prerendering, blender is still missing a lot of the professional/proprietaries tools that gives the last 10% of polish in big budget productions (photorealism). The community has done a great job in last 10 years but there's still a lot of technical tools locked behind something like max/maya ecosystems professional paywall that most people eventually transitions to for "serious" industry work because pipelines are hard to change. At least that's the state a few years ago.
Take a look at the earlier concepts / renderings for Shrek! Before the studio gutted the team(?) and told everyone to pull their heads in. Absolutely off-putting.
Readit News
"Proper" production renderers like Cycles do look better of course, but having an alternative which is viable on a shoestring budget is very valuable.
Also, having seen the film, I found the "unrealistic", cartoonish look very much to be a creative choice. Evee can produce much more "realistic" renders than what you see in the movie, but this requires also much more investment into things like assets and textures, otherwise you quickly land in the uncanny valley. So I think switching to Cycles probably would not matter much, unless the creators would also change their creative choices, which would result in a different movie, but not necessarily a better one.
"rules to create films based on the traditional values of story, acting, and theme, while excluding the use of elaborate special effects or technology. It was supposedly created as an attempt to "take back power for the directors as artists" as opposed to the movie studio."
https://en.wikipedia.org/wiki/Dogme_95
I had a negative initial reaction to the animation style but it hooked me in and blew me away. It had virtues far more vital than render quality. In contrast, I bailed on "Inside Out 2" and have no interest retrying. I hope more people are encouraged to create lofi meaningful movies instead of thinking it's the preserve of billion dollar studios and sweat-shop animation factories.
Like the first time I played Super Mario Bros. on an LED screen. Finally I could see each pixel clearly, exactly the way the original artist didn't intend!
Edit: in all seriousness, this makes me wonder: has anyone ever re-orchestrated Beethoven's Fifth? Say, in the orchestration style of Ravel or Strauss? Someone must have done this, even as a joke, and I'd love to hear it. (I know about the "Fifth of Beethoven" disco tune which is great, but that's not what I'm asking about.)
https://www.youtube.com/watch?app=desktop&v=a0GW8Na5CIE
Blender has a lot of other problems, but CUDA/Optix support is there for reasonable hardware =3
Last I heard that was the advantage of the propriety/in-house alternatives.
It will never be on-par with Eevee's performance though as they are fundamentally different approaches to rendering: Cycles is a physically-based path-tracing engine, while Eevee uses rasterization through OpenGL.
1: https://code.blender.org/2021/04/cycles-x/
The distant and "landscape" views look very nice, and in stark contrast to the game-like and amateur rendering of close up scenes with the animals. They don't even have anti-aliasing and the things look "blocky".
I hope this thing won because of the story and characters, and not its visuals.
The movie doesn’t look real, but it also doesn’t act real either.
Seems like a fluke, though.
The story is not, and the visuals are sufficient to tell the story.
Since they're not going crazy with effects it seems like a good compromise
Deleted Comment
But oh boy, what an amazing cutscene to watch. I'm worried that the story the media is putting forward is that this was an innovative and cutting edge movie - based only on a superficial appreciation of the (stunning) art design. But the real story is how the director worked within his limitations to make something equally enjoyable and meaningful as the other guys.
Most importantly, this movie passed the Actual Kid (TM) test. My 7 year old and his friends sat raptured through the entire movie without any slapstick, pop music numbers, or even dialogue! Not once, but 4 times now!
It's not about the textures and shadows.
From the more recent ones I highly recommend "Sprite Fright".
Several years ago I remember that after a year where the movie that won best animated was not the one that those in the animation industry overwhelming thought was sure to win some animation industry magazine survived Academy members asking which movie they voted for and why.
What they found was that a large number of the voters thought of animated movies as just for little kids and hadn't actually watched any of the nominees. They picked their vote by whatever they remembered children in their lives watching.
E.g., if they were parents of young children, they'd vote for whatever movie that their kids kept watching over and over. If they no longer had children at home they would ask grandkids or nieces or nephews "what cartoon did you like last year?" and vote for that.
Another factor was that a lot of these people would vote for the one they had heard the most about.
That gives Disney a big advantage. How the heck did Flow overcome that?
Inside Out 2 had a much wider theatrical release in the US, was widely advertised, made $650 million domestic, is the second highest grossing animated movie of all time so far worldwide, and streams on Disney+.
All that should contribute to making it likely that those large numbers of "vote even though they don't watch animated movies" Academy members would have heard of it.
Flow had a small US theatrical release at the end of the year. I didn't see any advertising for it. I'd expect a lot of Academy members hadn't heard of it.
As a guess, maybe Moana 2 is the movie that the kids are repeat streaming. That was not a nominee so maybe those "vote for what my kid watched" voters didn't vote this year and so we actually got a year where quality non-Disney movies had a chance?
* Gints Zilbalodis, who is 30 years old, is the youngest director to win the Oscar for best animated feature.
* Flow is the first fully-European produced and funded film to win the feture animation Oscar.
* Flow is the first dialogue-less film to win the feature animation Oscar.
* Flow, made for under $4 million, is by far the lowest-budget film to ever win the category.
It also says the winner of the animated short category, In the Shadow of the Cypress, was unexpected since the Iranian filmmakers couldn't do any of the usual in-person campaigning of Academy voters due to visa problems.
[1] https://www.cartoonbrew.com/awards/underdogs-win-latvias-flo...
1. The academy has had a significant increase of young voters in the past 10 years or so. Generally speaking, young voters are more likely to take animation as a "serious" medium.
2. These interviews were always somewhat overstated. Of course some voters have stupid rationales, but I don't think this dominates the academy.
3. Disney's Inside Out 2 was nowhere close to winning the award this year - Flow's biggest competition was The Wild Robot, which did gross far more than Inside Out 2, but far below Inside Out 2.
If you look at the past couple years, The Boy and the Heron (Studio Ghibli) won over Across the Spider-Verse (with Pixar's movie Elemental nowhere close) in 2023, Guillermo del Toro's Pinocchio won over Across the Spider-Verse (with Pixar's movie Turning Red nowhere close) in 2022, etc.
I'm curious what year you're thinking about above. Perhaps Toy Story 4 over Klaus in 2019?
Exactly the same as Inside Out 2 then?
(I'm guessing it was far more than Flow but less than Inside Out 2?)
The Oscars are the slowest possible reflection of social change, and I’m sure the perspective you share is still held my many members, but this win holds out some hope for sure.
I had a friend who was a Recording Academy member as a classical musician. He thought it was strange that they asked him to vote for the best hip-hop album since he doesn't listen to hip-hop at all.
So for many of the categories that are a little more niche, it basically turns into a popularity contest, rather than the opinion of true experts.
It had good ideas but didn't do very well with them (contrary to the first movie, which was great). I'm not surprised a movie which wasn't "just a sequel" managed to beat Moana and IO2.
The Academy has a reputation for seeking "artistic merit" even at a cost of good entertainment. They're hoping to advance something that didn't do well at the box office. Sometimes that means giving awards to films that turn out to be dogs, but sometimes they manage to promote things that deserve attention.
A lot of Oscar-bait gets a small release at the end of the year, to qualify it for the Oscars. If it gets a nomination, they'll use that as part of a wider campaign later. That's why they send out screeners: they know that many members won't have had a chance to see it in the theater.
With most media since the dawn of Hollywood, the internet and now AI, we are accustomed to being told exactly what is happening. Think about how 'laugh tracks' tell you to laugh. The search for an answer or meaning of something is largely taken away from you. Without that instruction you are left to make your own interpretation of things, no delivery of a specific message or theme. This means the movie is experienced differently by everyone. That why it's so great.
I remember doing all the tutorials when I was younger and considering game dev.
Example: https://images.squarespace-cdn.com/content/v1/58586fa5ebbd1a...
There are probably some flaws here as well, but you need to study the picture in detail. And Flow used the fast renderer of Blender, not the quality one.
Still, it does have a unique style that is much more interesting than many other animated movies. So what is technically impressive, just throwing more compute at it to make it photorealistic?
I think art style will have a larger impact. In a way it is technically impressive as it didn't need a lot of compute power.
Deleted Comment
We did come a long way
This movie was rendered with the latter.
Like any 3D package, you can also install other renderers.
So any perceived deficit in picture quality here is more to do with budget than some limitation of Blender.