Загрузка видео...

Не удалось загрузить видео

На главную

I've compiled some performance tips gleaned from working on my 1 million GPU AI steering agents project. I want to see Twitter compression destroy this video. 🚀 GPGPU: N-Body - O(n^2) - simulations are easily processed in parallel. GPUs are really good at parallel processing = 🫶. 🌐 Spatial...

TheMirzaBeig's profile picture

Mirza Beig

28,961 subscribers

13,252 просмотров • 2 лет назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

We figured you needed to see LiquiGen today. The simulation, rendering, and meshing are all real-time.
0:58

We figured you needed to see LiquiGen today. The simulation, rendering, and meshing are all real-time.

JangaFX Software

109,941 просмотров • 3 лет назад

WATER SIMULATION IS WORKING!!! It's extremely non-optimized, and there are still some glitches, and of course the rendering is very temporary, but IT'S WORKING! #indiedev #gamedev #indiegames
1:34

WATER SIMULATION IS WORKING!!! It's extremely non-optimized, and there are still some glitches, and of course the rendering is very temporary, but IT'S WORKING! #indiedev #gamedev #indiegames

Nikita Lisitsa

13,281 просмотров • 4 месяцев назад

Ok, I've decided to implement Particle Life simulation on GPU, using WebGPU, in browser. For now I've spawned 1'000'000 particles and implemented basic rendering & camera movement. I recon computing interaction forces will be the most painful part to implement...
0:20

Ok, I've decided to implement Particle Life simulation on GPU, using WebGPU, in browser. For now I've spawned 1'000'000 particles and implemented basic rendering & camera movement. I recon computing interaction forces will be the most painful part to implement...

Nikita Lisitsa

16,249 просмотров • 1 год назад

I wanted to see how far I could realise lapping beach water in #UnrealEngine, Simmed in #Houdini, Using UE5-Main (5.6) Raster rendering with the new Raytraced Translucency and Lumen. #UE5 #GameDev #Cinematic #VFX #Simulation #RayTracing
0:57

I wanted to see how far I could realise lapping beach water in #UnrealEngine, Simmed in #Houdini, Using UE5-Main (5.6) Raster rendering with the new Raytraced Translucency and Lumen. #UE5 #GameDev #Cinematic #VFX #Simulation #RayTracing

Dylan Browne

30,503 просмотров • 1 год назад

We're currently working on the endgame fight cinematic and need your help! We're looking for volunteers experienced in any of the following: lighting, rendering, set dressing, VFX arts, and/or destruction simulation Apply as a #3DArtist at:
0:25

We're currently working on the endgame fight cinematic and need your help! We're looking for volunteers experienced in any of the following: lighting, rendering, set dressing, VFX arts, and/or destruction simulation Apply as a #3DArtist at:

Skyblivion

14,343 просмотров • 7 месяцев назад

I am at the 4th iteration on the pixel inspector and light path debugging tool. Eventually I moved the path collection to the GPU, so I don't have to toggle between CPU and GPU rendering, making the visualization fully interactive.
1:47

I am at the 4th iteration on the pixel inspector and light path debugging tool. Eventually I moved the path collection to the GPU, so I don't have to toggle between CPU and GPU rendering, making the visualization fully interactive.

Max Liani

83,377 просмотров • 2 лет назад

📢Spinners for Three.js . Copy, paste, done. GPU-based point rendering, lightweight and flexible. No hard to get, no mystery. Based on the work of Paidax and Mike Bespalov You can also edit the parameters to achieve even more variations. Link below!
0:34

📢Spinners for Three.js . Copy, paste, done. GPU-based point rendering, lightweight and flexible. No hard to get, no mystery. Based on the work of Paidax and Mike Bespalov You can also edit the parameters to achieve even more variations. Link below!

𝗕𝗮𝗻𝗱𝗶𝗻𝗼𝗽𝗹𝗮

54,406 просмотров • 2 месяцев назад

I don't believe we live in a simulation like in the movie. I believe we are energy, but not a simulation. The video doesn't look real. But. Draw your own conclusions. My Spanish isn't very good, but this guy is saying something about the simulation failing with free energy, and that's why there are pyramids everywhere to generate free energy. Without cables and with a lot of free energy, the system can overload and the simulation ends in failure. I don't know exactly.
7:03

I don't believe we live in a simulation like in the movie. I believe we are energy, but not a simulation. The video doesn't look real. But. Draw your own conclusions. My Spanish isn't very good, but this guy is saying something about the simulation failing with free energy, and that's why there are pyramids everywhere to generate free energy. Without cables and with a lot of free energy, the system can overload and the simulation ends in failure. I don't know exactly.

Bronze Giant

456,269 просмотров • 10 месяцев назад

The performance of multi-viewport rendering is not different from the single-viewport mode in #Tellusim Engine. It allows using an enormous number of screens/HMDs to a single PC, rendering the same scene using multiple GPUs. A single 2080Ti can easily run 5 2k screens with Google Map Tiles API data.
0:42

The performance of multi-viewport rendering is not different from the single-viewport mode in #Tellusim Engine. It allows using an enormous number of screens/HMDs to a single PC, rendering the same scene using multiple GPUs. A single 2080Ti can easily run 5 2k screens with Google Map Tiles API data.

Alexander Zapryagaev

54,505 просмотров • 2 лет назад

The next frontier of AI is physical. 🤖 From simulation to synthetic data, developers are training robots to understand the real world with NVIDIA Cosmos, Isaac simulation frameworks and open-source physics engine, Newton. Watch the video to see how partners like Hon Hai Technology Group (Foxconn), Hexagon, Humanoid, Noble Machines, and Skild AI are building the future. 🎥
0:51

The next frontier of AI is physical. 🤖 From simulation to synthetic data, developers are training robots to understand the real world with NVIDIA Cosmos, Isaac simulation frameworks and open-source physics engine, Newton. Watch the video to see how partners like Hon Hai Technology Group (Foxconn), Hexagon, Humanoid, Noble Machines, and Skild AI are building the future. 🎥

NVIDIA Robotics

21,515 просмотров • 2 месяцев назад

testing my osu rendering in roblox with a 9 minute long XNOR XNOR XNOR osu aspire entry! there's still visual bugs i need to work out, especially with sliders, but i'm happy with the performance 🙂
1:15

testing my osu rendering in roblox with a 9 minute long XNOR XNOR XNOR osu aspire entry! there's still visual bugs i need to work out, especially with sliders, but i'm happy with the performance 🙂

littensy

22,595 просмотров • 1 год назад

For my 3d simulation I needed to implement new rendering. To replace "sphere par particle" with marching cubes. I never implemented that kind of stuff, so I decided to first try a simplified 2d version, the marching squares. It worked, very fast rendering now.
0:23

For my 3d simulation I needed to implement new rendering. To replace "sphere par particle" with marching cubes. I never implemented that kind of stuff, so I decided to first try a simplified 2d version, the marching squares. It worked, very fast rendering now.

Zolden

90,696 просмотров • 1 год назад

FASTER for #Amiga OCS/ECS: I never imagined saying this in a million years, but I seem to be struggling to get the performance I need out of the blitter, and I don't even have all the stuff on screen yet! Anybody know a faster way of rendering bobs than the cookie cutter method?
1:58

FASTER for #Amiga OCS/ECS: I never imagined saying this in a million years, but I seem to be struggling to get the performance I need out of the blitter, and I don't even have all the stuff on screen yet! Anybody know a faster way of rendering bobs than the cookie cutter method?

Jonathan Thomas @atariste.bsky.social

13,879 просмотров • 1 год назад

🌊: Ever wondered what cutting-edge physics simulation looks like in the browser? Check out saharan / さはら's collection at - Realistic water droplets with WebAssembly SIMD - Interactive fluffy ball sim using GPGPU - Coupled cloth and fluid dynamics in real-time - A digital clock powered by a custom physics engine Saharan built everything in Haxe using techniques like MLS-MPM and parallel computation. The physical accuracy and performance are exceptional. All MIT-licensed on GitHub. Perfect if you're exploring GPGPU techniques, browser physics, or just want to see what's possible with WebGL/WebGPU. #WebGL #WebGPU #Physics #GameDev #ComputationalPhysics
1:55

🌊: Ever wondered what cutting-edge physics simulation looks like in the browser? Check out saharan / さはら's collection at - Realistic water droplets with WebAssembly SIMD - Interactive fluffy ball sim using GPGPU - Coupled cloth and fluid dynamics in real-time - A digital clock powered by a custom physics engine Saharan built everything in Haxe using techniques like MLS-MPM and parallel computation. The physical accuracy and performance are exceptional. All MIT-licensed on GitHub. Perfect if you're exploring GPGPU techniques, browser physics, or just want to see what's possible with WebGL/WebGPU. #WebGL #WebGPU #Physics #GameDev #ComputationalPhysics

WebGL / WebGPU

14,778 просмотров • 5 месяцев назад

There's been a few cool updates recently. In particular, Rerun 0.33 released headless rendering. This, along with the Fable 5 release pushed me to work torwards making MAMMA realtime! I threw Fable at the problem, and it was able to take original implementation that was ~12 seconds / frame and get it all the way down to 40ms /frame, or nearly a 300x speedup 🏎️ How did I achieve this? TLDR: - Use rerun's headless rendering as supervision when optimizing - Save rrd file as test fixture to guide model optiziation with /goal - create an html artifact with headless rendering to provide detailed breakdown of what it did and how it actually looks like in the viewer There were a few critical bits to make sure that this ACTUALLY worked and that Fable didn't just cheat or delete something and declare victory. The first is that the original version used Rerun, this allowed us to save things to disk as an RRD file, meaning we could query the contents and use this as a sort of test fixture or golden artifact that held EXACTLY what all of the values should be. Then we can use this with /goal as a metric when doing the optimization to ensure there are no regressions. The second bit is the headless rendering, this gave us the ability to check that not only did the test fixture pass, but it also looked visually correct. This made a huge difference, and an awesome side affect of it is that we can use the headless rendering to create an implementations.html file. This gives a visual guide as to what the agent did (I walk through it in the video below) Along with this, we're working on an MCP server for rerun that allows full interactivity with the rerun viewer for your agent. So for example the agent can click, drag, move views, scroll timelines, ect. I used this to help the agent debug certain parts such as when the 2d sam masks didn't line up, or if the triangulated keypoints werent correctly matching with the optimized mesh. The agents could go, click into the view, scroll through the timeline and see where things went wrong. Fable + Headless Rendering + Rerun MCP == 300x speedup in less then a days work With these new tools, I'm planning on going back to my gaussian splatting implemntation and cleaning it up + making it fast!
1:33

There's been a few cool updates recently. In particular, Rerun 0.33 released headless rendering. This, along with the Fable 5 release pushed me to work torwards making MAMMA realtime! I threw Fable at the problem, and it was able to take original implementation that was ~12 seconds / frame and get it all the way down to 40ms /frame, or nearly a 300x speedup 🏎️ How did I achieve this? TLDR: - Use rerun's headless rendering as supervision when optimizing - Save rrd file as test fixture to guide model optiziation with /goal - create an html artifact with headless rendering to provide detailed breakdown of what it did and how it actually looks like in the viewer There were a few critical bits to make sure that this ACTUALLY worked and that Fable didn't just cheat or delete something and declare victory. The first is that the original version used Rerun, this allowed us to save things to disk as an RRD file, meaning we could query the contents and use this as a sort of test fixture or golden artifact that held EXACTLY what all of the values should be. Then we can use this with /goal as a metric when doing the optimization to ensure there are no regressions. The second bit is the headless rendering, this gave us the ability to check that not only did the test fixture pass, but it also looked visually correct. This made a huge difference, and an awesome side affect of it is that we can use the headless rendering to create an implementations.html file. This gives a visual guide as to what the agent did (I walk through it in the video below) Along with this, we're working on an MCP server for rerun that allows full interactivity with the rerun viewer for your agent. So for example the agent can click, drag, move views, scroll timelines, ect. I used this to help the agent debug certain parts such as when the 2d sam masks didn't line up, or if the triangulated keypoints werent correctly matching with the optimized mesh. The agents could go, click into the view, scroll through the timeline and see where things went wrong. Fable + Headless Rendering + Rerun MCP == 300x speedup in less then a days work With these new tools, I'm planning on going back to my gaussian splatting implemntation and cleaning it up + making it fast!

Pablo Vela

10,338 просмотров • 13 дней назад

#OctaneX was featured in today’s Apple keynote! Apple users can augment the massive increase in rendering power on the M4 with access to near unlimited high performance decentralized GPU compute power on The Render Network using Octane X.
0:32

#OctaneX was featured in today’s Apple keynote! Apple users can augment the massive increase in rendering power on the M4 with access to near unlimited high performance decentralized GPU compute power on The Render Network using Octane X.

The Render Network

536,966 просмотров • 2 лет назад

NEW RENDERING: Bena Road 🚄 Here's a visual rendering of a high-speed train passing over Bena Road on a viaduct in the Tehachapi Mountains south of Bakersfield. *This is an initial design subject to revision.
0:55

NEW RENDERING: Bena Road 🚄 Here's a visual rendering of a high-speed train passing over Bena Road on a viaduct in the Tehachapi Mountains south of Bakersfield. *This is an initial design subject to revision.

CA High-Speed Rail 🚄💨

19,026 просмотров • 1 год назад

Amiga OpenGL remote rendering. OpenGL commands get serialised on Amiga, sent over a TCP socket to a remote rendering server. That is Amiga 68k GLQuake (running in winuae) rendering on my PC RTX3080 GPU :)
0:28

Amiga OpenGL remote rendering. OpenGL commands get serialised on Amiga, sent over a TCP socket to a remote rendering server. That is Amiga 68k GLQuake (running in winuae) rendering on my PC RTX3080 GPU :)

Claude Schwarz

40,876 просмотров • 5 месяцев назад

AI for robotics has some monumental challenges, like high dimensionality and reasoning with the physical world. Ashok Elluswamy, AI lead at Tesla, thinks the hardest problem is Evaluation. - Loss isn't a perfect indicator of Policy Neural Network's quality. - Testing in the real world for all the long-tail failure cases is not possible. To solve the evaluation challenge, Tesla has developed a World Simulator Model. It is trained on state and action pairs, so for a given current camera feed and current action, the model generates the next frame in the cameras. Once the world simulation is good enough, it can be hooked up to the policy network for a closed-loop simulation. In this setup, the model generates the next video frames, and based on those, the policy network comes up with the next action, which feeds back into the world simulation; this continuous loop creates a coherent simulation. The world simulation neural network is not just applicable to self-driving but also to humanoids, because Tesla trains it on common data. The examples show Optimus navigating or manipulating inside the world simulation, with all the pixels being generated by the world simulation model. The robotic agents can be tested in an accurately reproduced edge case and in other potential variations of the same edge case.
1:29

AI for robotics has some monumental challenges, like high dimensionality and reasoning with the physical world. Ashok Elluswamy, AI lead at Tesla, thinks the hardest problem is Evaluation. - Loss isn't a perfect indicator of Policy Neural Network's quality. - Testing in the real world for all the long-tail failure cases is not possible. To solve the evaluation challenge, Tesla has developed a World Simulator Model. It is trained on state and action pairs, so for a given current camera feed and current action, the model generates the next frame in the cameras. Once the world simulation is good enough, it can be hooked up to the policy network for a closed-loop simulation. In this setup, the model generates the next video frames, and based on those, the policy network comes up with the next action, which feeds back into the world simulation; this continuous loop creates a coherent simulation. The world simulation neural network is not just applicable to self-driving but also to humanoids, because Tesla trains it on common data. The examples show Optimus navigating or manipulating inside the world simulation, with all the pixels being generated by the world simulation model. The robotic agents can be tested in an accurately reproduced edge case and in other potential variations of the same edge case.

The Humanoid Hub

16,585 просмотров • 4 месяцев назад