Jake Rice

Jake Rice

@tearsofjake.bsky.social

Computer Graphics Investigator (CGI)

295 followers74 following72 posts

Jake Rice
@tearsofjake.bsky.social

Wow dude your new Claude made SDF modeler is really sick. Really groundbreaking stuff. Can't wait to see what your crazy brain cooks up next!

Jake Rice
@tearsofjake.bsky.social

went to the Syd Mead exhibit at the Transamerica building. seeing his work in person is actually unbelievable. he was truly a master of color and design. his technical abilities as an artist and his amazing taste will probably never be matched again. just a true titan. if you get the chance, go!

Jake Rice
@tearsofjake.bsky.social

VoroNERF is coming along better now that i fixed a huge bug with my Top K selection of sites! It's still not perfect of course, lots of bugs and stuff to fix. This is with 60k voronoi sites, using the 16 closest sites around any sampling position along the render rays.

Jake Rice reposted

MOPs
@motionoperators.com

My MOPs course on Houdini.School is now free! If you want a start to finish course on how (almost) everything works, plus the math background, you can watch it all here: www.houdini.school/courses/mops...

houdini.schoolMOPs: Motion Operators for HoudiniThis course provides an overview of the MOPs workflow and how it integrates with the rest of Houdini, shows concrete examples of how MOPs can be used in both motion graphics and visual effects workflo...
Jake Rice
@tearsofjake.bsky.social

voronoi based NERF training isn't going incredibly well... 😂20k iterations with 80k voronoi sites takes about 10 hours to train. And it still looks dog water. Definitely need more sites and some better approach to accelerating sampling....

Jake Rice
@tearsofjake.bsky.social

Voronoi implicit but this time in 3d. It's not fully NERF mode yet, as I'm not doing any kind of sampling along rays, this is just trying to learn volumetric data. 30,000 Voronoi sites. Probably not enough for the pighead, but it's just a fun first test.

Jake Rice
@tearsofjake.bsky.social

I'm once again making neural implicits of my cats. This time we're back to the voronoi. Gabor noise style. We weight the result of the softmax at any site i by sin(F_i*(x_i-x_j)•u_i) where F_i is learned frequency and u_i is a learned anisotropy direction. I call it Gaboroni

Jake Rice
@tearsofjake.bsky.social

Similar to this post: bsky.app/profile/tear... we're using 3000 voronoi sites. It captures the high frequency info worse than the anisotropic voronoi which is interesting. But it's also one less parameter to learn.

Jake Rice@tearsofjake.bsky.social

I had the idea to incorporate anisotropy into the recent Spherical Voronoi paper. And when we apply their ideas to euclidean problems (not Spherical) the results are pretty great when anisotropy is used. This is 3000 anisotropic sites vs 3000 isotropic sites. Same learning rates for both

Jake Rice
@tearsofjake.bsky.social

Wanted to make a neural implicit that's just layers of anisotropic simplex noise. Turns out it works pretty well. With 24 layers of simplex noise each with a 96x96 texture of anisotropy data we can get this kind of result! It's not at all a good compression method but it's fun and cool :)

Jake Rice
@tearsofjake.bsky.social

These are the noise layers that are all summed up to get the predicted image. each noise layer is colored with two colors, and each layer has a circular mask associated with it as well. I scaled the values here by a factor of ten so that they're easier to see.

Jake Rice
@tearsofjake.bsky.social

I had the idea to incorporate anisotropy into the recent Spherical Voronoi paper. And when we apply their ideas to euclidean problems (not Spherical) the results are pretty great when anisotropy is used. This is 3000 anisotropic sites vs 3000 isotropic sites. Same learning rates for both

Jake Rice
@tearsofjake.bsky.social

Shockingly annoying to find the points of intersection between a plane and an AABB. Even when I know the plane passes through the center of the box. If anyone has a simple method for finding the exact distance to a plane clipped by a bounding box I'd be eternally grateful.

Jake Rice reposted

Baptiste Genest
@baptiste-genest.bsky.social

Computing the exact bijection of the optimal transport (OT) problem between very large point sets is completely untractable… In our SIGGRAPH Asia 2025 paper: “BSP-OT: Sparse transport plans between discrete measures in log-linear time” we get one with typically 1% of error in a few seconds on CPU!

Jake Rice
@tearsofjake.bsky.social

For the snow on the hedge maze in Zootopia 2, we cast rays upwards from each leaf and if it didn't collide with anything we instanced a snow chunk onto it. We made a few variants of the snow chunks to add extra variation.

Nikita Lisitsa@lisyarus.bsky.social

Light-snow covered short grass has some unexpectedly intricate texture to it

Jake Rice
@tearsofjake.bsky.social

I'd imagine snow covered grass is relatively similar, but on its side. Each blade could have a wind direction ray and if the ray doesn't hit any other blade of grass in the wind direction (over a certain distance), then it could be snow covered. And then combo that with some general snow coverage

Jake Rice
@tearsofjake.bsky.social

my FX work alone required at least 1 quadrillion of the 3.7 quadrillion rays traced on Zootopia 2. This is FACTUAL and REAL. If you are a journalist, please cite this in your article.

Yining Karl Li@yiningkarlli.bsky.social

Using some very approximate guesstimate math, I think we traced around 3.7 quadrillion individual rays to produce the final movie. This does not include intermediate / WIP renders. I think this is astronomy-accurate, meaning it's in roughly the correct order of magnitude. (2/2)

View full thread
Jake Rice
@tearsofjake.bsky.social

The least obvious part of all of this is how to construct that metric tensor G from a given vector, but it turns out it's super straight forward. The attached example shows how to do so for a 3D metric tensor. In our paper we outline specific bounds for the sizing of this metric.

Jake Rice
@tearsofjake.bsky.social

Just to elucidate the idea a bit more, on the left we have the original perlin noise algorithm, and on the right we have the steerable version, with the changes underlined in red. We simply add a second set of weights, and incorporate our metric into the original dot product step

Jake Rice reposted

Yining Karl Li
@yiningkarlli.bsky.social

One of my coworkers, @tearsofjake.bsky.social, has a talk at SIGGRAPH this year about this really cool steerable perlin noise technique that was used on Moana 2. He's just posted some handy reference implementations for Houdini, Unity, Godot, and Blender; check it out! github.com/jakericedesi...

github.comGitHub - jakericedesigns/SteerablePerlinNoise: Implementations of "Steerable Perlin Noise" as presented at Siggraph 2025Implementations of "Steerable Perlin Noise" as presented at Siggraph 2025 - jakericedesigns/SteerablePerlinNoise
Jake Rice
@tearsofjake.bsky.social

How do I get the discover page to align more with my interests? I've tried following a bunch of people, but it still kinda just shows stuff based off whatever categories I clicked when I signed up...

Jake Rice reposted

Matt Ebb
@mattebb.com

I know it’s a bit uncool to linkedin-post on here but we’re hiring at work, I’ve been looking at quite a lot of showreels, so I wrote up some tips for making a great fx reel #vfx www.linkedin.com/pulse/creati...

linkedin.comCreating a stand-out Effects reelAs a VFX artist, your showreel is one of the most important tools in the job search process. Whether you’re trying to break into the industry or looking for a new role, having a stand-out showreel can...

Jake Rice reposted

Morten Vassvik
@vassvik.bsky.social

So what is the tiniest possible fluid simulation? Can we get anything interesting from a single cell? For this we'll be using a standard MAC grid, which means we'll represent horizontal velocities (red) on the vertical edges and vertical velocities (green) on the horizontal edges of each cell.

Jake Rice
@tearsofjake.bsky.social

If you were curious why I was after the intersections of paraboloids, this is why! :) bsky.app/profile/tear...

Jake Rice@tearsofjake.bsky.social

Is there a robust and fast way of finding the nearest point on a paraboloid of form: (x,y, f(x) + f(y)) where f() could be any range of convex functions, like x^2 or x^4? So like given a point (p0,p1,p2) minimize: (x - p0)^2 + (y-p1)^2 + ((f(x) + f(y)) - p2)^2 With respect to x and y.

Jake Rice reposted

Yining Karl Li
@yiningkarlli.bsky.social

Recordings for Graphics Programming Conference 2024 are now up. The lineup is packed with amazing talks ranging from rendering for games like Tiny Glade, Hades, and Baldur’s Gate 3, to even a talk on designing your own GPU! youtube.com/playlist?lis...

youtube.comGraphics Programming Conference 2024 - YouTubeThe Graphics Programming Conference is a three day event in November. We have keynotes from industry experts, presentations, and “masterclass” workshops with...
Jake Rice
@tearsofjake.bsky.social

Is there a robust and fast way of finding the nearest point on a paraboloid of form: (x,y, f(x) + f(y)) where f() could be any range of convex functions, like x^2 or x^4? So like given a point (p0,p1,p2) minimize: (x - p0)^2 + (y-p1)^2 + ((f(x) + f(y)) - p2)^2 With respect to x and y.

You are up to date.