Bart Wronski 🇺🇦🇵🇸
@bartwr.bsky.social
Engineering, Computer Graphics, Art, DSP, ML
Culture, Techno, Industrial, and Electronic Music.
Research Scientist at NVIDIA.
Ex Google Research, Ex games (Sony, Ubisoft, CD Projekt).
Politically leftist. He/they.
https://linktr.ee/bartwronsk
Culture, Techno, Industrial, and Electronic Music.
Research Scientist at NVIDIA.
Ex Google Research, Ex games (Sony, Ubisoft, CD Projekt).
Politically leftist. He/they.
https://linktr.ee/bartwronsk
Reposted by Bart Wronski 🇺🇦🇵🇸
I highly recommend watching this segment, not just because the CBS News execs and the White House didn’t want you to, but because these men were tortured and they deserved to have their voices heard.
!! Here’s a link to full video of the 60 Minutes segment that Bari Weiss killed last minute, via @jasonparis.bsky.social:
is.gd/paU8Ko
(It was uploaded to the Global TV app in Canada, seemingly by accident, and has now been taken down)
is.gd/paU8Ko
(It was uploaded to the Global TV app in Canada, seemingly by accident, and has now been taken down)
December 22, 2025 at 10:46 PM
I highly recommend watching this segment, not just because the CBS News execs and the White House didn’t want you to, but because these men were tortured and they deserved to have their voices heard.
Reposted by Bart Wronski 🇺🇦🇵🇸
I got the first post about direct lighting material occlusion up on my blog. On this one I go over the commonly used micro-occlusion approach and its limitations, and I start digging into micro-shadowing as an improvement.
irradiance.ca/posts/micros...
irradiance.ca/posts/micros...
December 20, 2025 at 9:31 PM
I got the first post about direct lighting material occlusion up on my blog. On this one I go over the commonly used micro-occlusion approach and its limitations, and I start digging into micro-shadowing as an improvement.
irradiance.ca/posts/micros...
irradiance.ca/posts/micros...
Reposted by Bart Wronski 🇺🇦🇵🇸
Legit super excited about this work coming out. My amazing doctoral student @ben.graphics has been working on an idea to use physically based differentiable rendering (PBDR) to probe visual understanding. Here, we generate physically-grounded metamers for vision models. 1/4
arxiv.org/abs/2512.12307
arxiv.org/abs/2512.12307
December 17, 2025 at 9:17 PM
Legit super excited about this work coming out. My amazing doctoral student @ben.graphics has been working on an idea to use physically based differentiable rendering (PBDR) to probe visual understanding. Here, we generate physically-grounded metamers for vision models. 1/4
arxiv.org/abs/2512.12307
arxiv.org/abs/2512.12307
Reposted by Bart Wronski 🇺🇦🇵🇸
My "No Graphics API" blog post is live! Please repost :)
www.sebastianaaltonen.com/blog/no-grap...
I spend 1.5 years doing this. Full rewrite last summer and another partial rewrite last month. As Hemingway said: "First draft of everything is always shit".
www.sebastianaaltonen.com/blog/no-grap...
I spend 1.5 years doing this. Full rewrite last summer and another partial rewrite last month. As Hemingway said: "First draft of everything is always shit".
No Graphics API — Sebastian Aaltonen
Graphics APIs and shader languages have significantly increased in complexity over the past decade. It’s time to start discussing how to strip down the abstractions to simplify development, improve pe...
www.sebastianaaltonen.com
December 16, 2025 at 6:52 PM
My "No Graphics API" blog post is live! Please repost :)
www.sebastianaaltonen.com/blog/no-grap...
I spend 1.5 years doing this. Full rewrite last summer and another partial rewrite last month. As Hemingway said: "First draft of everything is always shit".
www.sebastianaaltonen.com/blog/no-grap...
I spend 1.5 years doing this. Full rewrite last summer and another partial rewrite last month. As Hemingway said: "First draft of everything is always shit".
Reposted by Bart Wronski 🇺🇦🇵🇸
Super happy and honored to share that our paper "BSP-OT: Sparse transport plans between discrete measures in log-linear time" won a *Best paper award* at SIGGRAPH Asia 2025!
If you are here, come see my presentation about this work Wednesday afternoon!
Many thanks to the award committee!
If you are here, come see my presentation about this work Wednesday afternoon!
Many thanks to the award committee!
December 15, 2025 at 3:31 AM
Super happy and honored to share that our paper "BSP-OT: Sparse transport plans between discrete measures in log-linear time" won a *Best paper award* at SIGGRAPH Asia 2025!
If you are here, come see my presentation about this work Wednesday afternoon!
Many thanks to the award committee!
If you are here, come see my presentation about this work Wednesday afternoon!
Many thanks to the award committee!
Reposted by Bart Wronski 🇺🇦🇵🇸
Our #SIGGRAPHAsia 2025 paper "Sample Space Partitioning and Spatiotemporal Resampling for Specular Manifold Sampling" improves #realtime #caustics rendering.
Project Page: graphics.cs.utah.edu/research/pro...
@daqilin.bsky.social @tizian.bsky.social
Project Page: graphics.cs.utah.edu/research/pro...
@daqilin.bsky.social @tizian.bsky.social
December 14, 2025 at 10:12 PM
Our #SIGGRAPHAsia 2025 paper "Sample Space Partitioning and Spatiotemporal Resampling for Specular Manifold Sampling" improves #realtime #caustics rendering.
Project Page: graphics.cs.utah.edu/research/pro...
@daqilin.bsky.social @tizian.bsky.social
Project Page: graphics.cs.utah.edu/research/pro...
@daqilin.bsky.social @tizian.bsky.social
Reposted by Bart Wronski 🇺🇦🇵🇸
December 12, 2025 at 11:58 PM
Reposted by Bart Wronski 🇺🇦🇵🇸
Blog post on motion blur rendering (which is impressively thorough) by Alex Gauggel
gaukler.github.io/2025/12/09/n...
gaukler.github.io/2025/12/09/n...
Notes On Motion Blur Rendering
Introduction
gaukler.github.io
December 10, 2025 at 10:34 PM
Blog post on motion blur rendering (which is impressively thorough) by Alex Gauggel
gaukler.github.io/2025/12/09/n...
gaukler.github.io/2025/12/09/n...
My wife is very technical (software engineer) but has zero experience in computer graphics or AI, and yet she immediately, within a second, recognizes the "GenAI look" and says it looks like crap. I hope it's a widespread reaction. ¯\_(ツ)_/¯
McDonald's has released an AI-generated Christmas ad
The studio behind it says they 'hardly slept' for several weeks while writing AI prompts and refining the shots — 'AI didn't make this film. We did'
Comments have been turned off on YouTube
The studio behind it says they 'hardly slept' for several weeks while writing AI prompts and refining the shots — 'AI didn't make this film. We did'
Comments have been turned off on YouTube
December 9, 2025 at 3:38 PM
My wife is very technical (software engineer) but has zero experience in computer graphics or AI, and yet she immediately, within a second, recognizes the "GenAI look" and says it looks like crap. I hope it's a widespread reaction. ¯\_(ツ)_/¯
Reposted by Bart Wronski 🇺🇦🇵🇸
Some random Morton Code tricks:
If you've worked with Morton codes you've probably seen something like the this GLSL code to decode a Morton code, which end up at roughly 27 integer operations to generate the equivalent 2D position.
Code transcribed from fgiesen.wordpress.com/2009/12/13/d...
If you've worked with Morton codes you've probably seen something like the this GLSL code to decode a Morton code, which end up at roughly 27 integer operations to generate the equivalent 2D position.
Code transcribed from fgiesen.wordpress.com/2009/12/13/d...
December 5, 2025 at 2:07 PM
Some random Morton Code tricks:
If you've worked with Morton codes you've probably seen something like the this GLSL code to decode a Morton code, which end up at roughly 27 integer operations to generate the equivalent 2D position.
Code transcribed from fgiesen.wordpress.com/2009/12/13/d...
If you've worked with Morton codes you've probably seen something like the this GLSL code to decode a Morton code, which end up at roughly 27 integer operations to generate the equivalent 2D position.
Code transcribed from fgiesen.wordpress.com/2009/12/13/d...
Reposted by Bart Wronski 🇺🇦🇵🇸
Hi this is Beatrix ^^ dunno what I’m doing here… or anywhere really. My stuff:
- Artemis the cat and Lizzy the leopard gecko
- crocheting and other handcraft
- gaming, horror, horror games
- photography
- future graphics programmer burnout
- yeah
#introductions
- Artemis the cat and Lizzy the leopard gecko
- crocheting and other handcraft
- gaming, horror, horror games
- photography
- future graphics programmer burnout
- yeah
#introductions
December 5, 2025 at 1:17 PM
Hi this is Beatrix ^^ dunno what I’m doing here… or anywhere really. My stuff:
- Artemis the cat and Lizzy the leopard gecko
- crocheting and other handcraft
- gaming, horror, horror games
- photography
- future graphics programmer burnout
- yeah
#introductions
- Artemis the cat and Lizzy the leopard gecko
- crocheting and other handcraft
- gaming, horror, horror games
- photography
- future graphics programmer burnout
- yeah
#introductions
What a cool paper: arxiv.org/abs/2512.02371
Augmenting Halide language to support scheduling that allows for the use of GPU Tensor Cores (and similar) for non-ML, general image/audio processing applications.
IMO, shading languages badly need "scheduling" and heterogeneity of devices/passes.
Augmenting Halide language to support scheduling that allows for the use of GPU Tensor Cores (and similar) for non-ML, general image/audio processing applications.
IMO, shading languages badly need "scheduling" and heterogeneity of devices/passes.
Pushing Tensor Accelerators Beyond MatMul in a User-Schedulable Language
Tensor accelerators now represent a growing share of compute resources in modern CPUs and GPUs. However, they are hard to program, leading developers to use vendor-provided kernel libraries that suppo...
arxiv.org
December 4, 2025 at 3:11 AM
What a cool paper: arxiv.org/abs/2512.02371
Augmenting Halide language to support scheduling that allows for the use of GPU Tensor Cores (and similar) for non-ML, general image/audio processing applications.
IMO, shading languages badly need "scheduling" and heterogeneity of devices/passes.
Augmenting Halide language to support scheduling that allows for the use of GPU Tensor Cores (and similar) for non-ML, general image/audio processing applications.
IMO, shading languages badly need "scheduling" and heterogeneity of devices/passes.
Reposted by Bart Wronski 🇺🇦🇵🇸
🌶️ Hot take: ML-based graphics makes some graphics researchers unhappy for the same reason diffusion models make artists unhappy.
Both automate the interesting and beautiful parts of the work and both deskill the worker. Both result in questionable quality and both are being adopted regardless.
Both automate the interesting and beautiful parts of the work and both deskill the worker. Both result in questionable quality and both are being adopted regardless.
The best quote I’ve heard about AI was a half-joke by a coworker: “is there anything that classical techniques can do that AI can’t pretend to do better?” Sometimes pretending is good enough and sometimes it’s really not! That seems largely lost in the pro/anti AI discourse.
December 2, 2025 at 5:32 PM
🌶️ Hot take: ML-based graphics makes some graphics researchers unhappy for the same reason diffusion models make artists unhappy.
Both automate the interesting and beautiful parts of the work and both deskill the worker. Both result in questionable quality and both are being adopted regardless.
Both automate the interesting and beautiful parts of the work and both deskill the worker. Both result in questionable quality and both are being adopted regardless.
Reposted by Bart Wronski 🇺🇦🇵🇸
🧠🔬 Excited to share our #NeurIPS2025 paper: "Convolution Goes Higher-Order"!
We asked: Can shallow networks be as expressive as deep ones? Inspired by biological vision, we introduce higher-order convolutions that capture complex image patterns standard CNNs miss.
🧵👇
We asked: Can shallow networks be as expressive as deep ones? Inspired by biological vision, we introduce higher-order convolutions that capture complex image patterns standard CNNs miss.
🧵👇
December 1, 2025 at 1:24 PM
🧠🔬 Excited to share our #NeurIPS2025 paper: "Convolution Goes Higher-Order"!
We asked: Can shallow networks be as expressive as deep ones? Inspired by biological vision, we introduce higher-order convolutions that capture complex image patterns standard CNNs miss.
🧵👇
We asked: Can shallow networks be as expressive as deep ones? Inspired by biological vision, we introduce higher-order convolutions that capture complex image patterns standard CNNs miss.
🧵👇
For me, vibe-coding marked a new era of personal "tooling" and UX. Instead of searching for existing tools that are either too powerful/complex, or not an exact fit, I quickly hack together simple UX apps and background services for my use - using tech I wouldn't have time to learn otherwise.
December 1, 2025 at 3:20 AM
For me, vibe-coding marked a new era of personal "tooling" and UX. Instead of searching for existing tools that are either too powerful/complex, or not an exact fit, I quickly hack together simple UX apps and background services for my use - using tech I wouldn't have time to learn otherwise.
This is hilarious and a testament to Elon's manipulation and megalomania, but shows why no private and "closed" AI models should be used for any public systems and decision making.
We need: Open source, open weights, open datasets, and reproducible training.
And this is why they are fighting them.
We need: Open source, open weights, open datasets, and reproducible training.
And this is why they are fighting them.
November 21, 2025 at 2:11 AM
This is hilarious and a testament to Elon's manipulation and megalomania, but shows why no private and "closed" AI models should be used for any public systems and decision making.
We need: Open source, open weights, open datasets, and reproducible training.
And this is why they are fighting them.
We need: Open source, open weights, open datasets, and reproducible training.
And this is why they are fighting them.
Them: this is more of a comment than a question, but...
Me:
Me:
November 15, 2025 at 8:31 PM
Them: this is more of a comment than a question, but...
Me:
Me:
Reposted by Bart Wronski 🇺🇦🇵🇸
Here is an alpha release of Fabric, my open source node based creative coding & prototyping env.
github.com/Fabric-Proje...
Enjoyed Quartz Composer, or want a more intuitive node based tool? Check it out, & help build w me!
#swift, #swiftui #metal, #realtime, #gpu #graphics
github.com/Fabric-Proje...
Enjoyed Quartz Composer, or want a more intuitive node based tool? Check it out, & help build w me!
#swift, #swiftui #metal, #realtime, #gpu #graphics
Releases · Fabric-Project/Fabric
Node Creative Coding / 3D / Image Processing tool inspired by Quartz Composer - Fabric-Project/Fabric
github.com
November 14, 2025 at 10:27 PM
Hostages. People held without any charge and released in exchanges etc. are called hostages. (And they are held in underground "tunnels".)
November 8, 2025 at 6:44 PM
Hostages. People held without any charge and released in exchanges etc. are called hostages. (And they are held in underground "tunnels".)
Two weeks ago I read some hate on Hasan Piker related to Zohran and how their association will sink his election results (lol). So I listened to 20-50min clips from his YT channel.
That dude is legit good: charismatic, nuanced, pragmatic, emphatic, and policing his audience against hate.
That dude is legit good: charismatic, nuanced, pragmatic, emphatic, and policing his audience against hate.
November 7, 2025 at 1:22 PM
Two weeks ago I read some hate on Hasan Piker related to Zohran and how their association will sink his election results (lol). So I listened to 20-50min clips from his YT channel.
That dude is legit good: charismatic, nuanced, pragmatic, emphatic, and policing his audience against hate.
That dude is legit good: charismatic, nuanced, pragmatic, emphatic, and policing his audience against hate.
TIL that Norbert Wiener, creator of cybernetics and one of the biggest names in early computer science, statistics, robotics, and control systems - was a workers' rights supporter, argued for labor involvement in tech, and became an anti-war activist after WW2, refusing any military-related funding.
November 2, 2025 at 12:01 AM
TIL that Norbert Wiener, creator of cybernetics and one of the biggest names in early computer science, statistics, robotics, and control systems - was a workers' rights supporter, argued for labor involvement in tech, and became an anti-war activist after WW2, refusing any military-related funding.
Reposted by Bart Wronski 🇺🇦🇵🇸
i had a chat with lukasz and @antovsky about gamedev and stuff, if youre into that sort of thing! the middle hour is the juiciest but its all fantastic, ofc ;) youtu.be/mFBmoCv5EcQ?...
Game Architectures with Alex Evans and Anton Mikhailov
YouTube video by Wookash Podcast
youtu.be
November 1, 2025 at 7:06 PM
i had a chat with lukasz and @antovsky about gamedev and stuff, if youre into that sort of thing! the middle hour is the juiciest but its all fantastic, ofc ;) youtu.be/mFBmoCv5EcQ?...
Reposted by Bart Wronski 🇺🇦🇵🇸
I gave a talk at Pacific Graphics 2025 on the topic of "Classical Computer Graphics in the Age of Generative AI". I've uploaded the recording to Youtube today.
www.youtube.com/watch?v=Vyci...
www.youtube.com/watch?v=Vyci...
"Classical Computer Graphics in the Age of Generative AI", Pacific Graphics 2025 Keynote, Tzu-Mao Li
YouTube video by Tzu-Mao Li
www.youtube.com
October 30, 2025 at 7:53 PM
I gave a talk at Pacific Graphics 2025 on the topic of "Classical Computer Graphics in the Age of Generative AI". I've uploaded the recording to Youtube today.
www.youtube.com/watch?v=Vyci...
www.youtube.com/watch?v=Vyci...
Reposted by Bart Wronski 🇺🇦🇵🇸
Mamdani's big secret:
www.kenklippenstein.com/p/mamdanis-b...
www.kenklippenstein.com/p/mamdanis-b...
Mamdani's Big Secret
How the impossible makes his ascent possible
www.kenklippenstein.com
October 30, 2025 at 12:02 AM
Mamdani's big secret:
www.kenklippenstein.com/p/mamdanis-b...
www.kenklippenstein.com/p/mamdanis-b...
TIL about Python 3.13's cute copy.replace() for creating modified copies of immutable (not only, but for them it's super convenient) types.
Point = namedtuple('Point', ['x', 'y', 'z'])
p0 = Point(x=0, y=50, z=10)
p1 = copy.replace(p0, x=5)
You can override __replace__ in your custom class too!
Point = namedtuple('Point', ['x', 'y', 'z'])
p0 = Point(x=0, y=50, z=10)
p1 = copy.replace(p0, x=5)
You can override __replace__ in your custom class too!
October 28, 2025 at 5:56 PM
TIL about Python 3.13's cute copy.replace() for creating modified copies of immutable (not only, but for them it's super convenient) types.
Point = namedtuple('Point', ['x', 'y', 'z'])
p0 = Point(x=0, y=50, z=10)
p1 = copy.replace(p0, x=5)
You can override __replace__ in your custom class too!
Point = namedtuple('Point', ['x', 'y', 'z'])
p0 = Point(x=0, y=50, z=10)
p1 = copy.replace(p0, x=5)
You can override __replace__ in your custom class too!