Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

Jolt multi-threaded is crazy🤯 Stresstest using our new ECS library, Three.js BatchedMesh and Jolt Physics. 1000+ dynamic bodies, 120 fps on a low-end PC while recording. But try it for yourself! Demo Link + Code in the comments. #gamedev #webgamedev

Brian Breiholz

1,518 subscribers

17,993 görüntüleme • 1 yıl önce •via X (Twitter)

Bilim & Teknoloji Oyun #gamedev #webgamedev

Anya Rossi• Live Now

Private livecam show

8 Yorum

Brian Breiholz profil fotoğrafı

Brian Breiholz1 yıl önce

Live Demo on Stackblitz: @pmndrs ECS libary: Demo Code:

Brian Breiholz profil fotoğrafı

Brian Breiholz1 yıl önce

cc @bruno_simon re: why use Jolt lots of taxing setups in this demo (i.e. much potential for tunneling, small gaps, kinematic + dynamic bodies) and of course the performance 🤌

Brian Breiholz profil fotoğrafı

Brian Breiholz1 yıl önce

Oh and I forget to tag @jrouwe 🙈 who not only wrote the Jolt Physics library but also maintains the JS port 🙌

Robert Borghesi profil fotoğrafı

Robert Borghesi1 yıl önce

@threejs 😱😱😱😱😱😱

Ademola profil fotoğrafı

Ademola1 yıl önce

@threejs Insane 🔥

Andreas Longva profil fotoğrafı

Andreas Longva1 yıl önce

@threejs Neat. Do you know if there's a high-level write-up on the algorithms used in Jolt?

Brian Breiholz profil fotoğrafı

Brian Breiholz1 yıl önce

@threejs Thanks! Yeah the native Jolt docs have quite a bit of high-level/general Info:

Ertuğrul Çetin profil fotoğrafı

Ertuğrul Çetin1 yıl önce

@threejs That would be great if Havok had that in @babylonjs

Benzer Videolar

I've been working on getting my C multicore-runtime to run on the web. it's still early, but this is a fully multi-threaded renderer + physics running on WebGPU and WASM. 100k cubes, 60 FPS I'll be benchmarking it against Unity ECS soon

I've been working on getting my C multicore-runtime to run on the web. it's still early, but this is a fully multi-threaded renderer + physics running on WebGPU and WASM. 100k cubes, 60 FPS I'll be benchmarking it against Unity ECS soon

Gabriel Dechichi

2,444,467 görüntüleme • 7 ay önce

Trying to fix the FPS issues today If I'm on MacBook Pro M3 with battery >20% or power connected I get 120 FPS The moment I go below <20% battery, M3 goes in power save mode I think so FPS drops to 30 FPS 30 FPS is fine but it also decreases the entire game speed by 30/120 (so 4x) Same on iPhone, I get 30 FPS, but the game gets too slow My Cursor written animate() function is, which runs on requestAnimationFrame, but I guess that deltaTime code needs some fixes to disconnect it from the FPS, I already tried but didn't work Ideally 120 FPS or 30 FPS or 10 FPS all runs at same speed even if lower or higher frame rate function animate() { requestAnimationFrame(animate); // Calculate delta time in seconds, but use fixed timestep for physics const currentTime = const rawDeltaTime = (currentTime - lastFrameTime) / 1000; deltaTime = 1/60; // Force fixed 60 FPS timestep for all physics lastFrameTime = currentTime;

Trying to fix the FPS issues today If I'm on MacBook Pro M3 with battery >20% or power connected I get 120 FPS The moment I go below <20% battery, M3 goes in power save mode I think so FPS drops to 30 FPS 30 FPS is fine but it also decreases the entire game speed by 30/120 (so 4x) Same on iPhone, I get 30 FPS, but the game gets too slow My Cursor written animate() function is, which runs on requestAnimationFrame, but I guess that deltaTime code needs some fixes to disconnect it from the FPS, I already tried but didn't work Ideally 120 FPS or 30 FPS or 10 FPS all runs at same speed even if lower or higher frame rate function animate() { requestAnimationFrame(animate); // Calculate delta time in seconds, but use fixed timestep for physics const currentTime = const rawDeltaTime = (currentTime - lastFrameTime) / 1000; deltaTime = 1/60; // Force fixed 60 FPS timestep for all physics lastFrameTime = currentTime;

@levelsio

446,331 görüntüleme • 1 yıl önce

Few know this yet, but Hyperfy has built an insanely well-performing WebXR engine. 🔹 Almost 100Hz feel on my Quest 3 🔹 Runs buttery smooth in a browser, standalone 🔹 Even screen recording doesn’t cause noticeable frame drops Hard to believe this is WebXR. Three.js #WebXR 🔗 Link to try it yourself in the comments!

Few know this yet, but Hyperfy has built an insanely well-performing WebXR engine. 🔹 Almost 100Hz feel on my Quest 3 🔹 Runs buttery smooth in a browser, standalone 🔹 Even screen recording doesn’t cause noticeable frame drops Hard to believe this is WebXR. Three.js #WebXR 🔗 Link to try it yourself in the comments!

𝚟𝚘𝚡𝚟𝚒𝚎𝚗𝚗𝚎

19,750 görüntüleme • 1 yıl önce

🤯 Just build Fortnite in React ... With the new releases of r3/timeline and r3/viverse, building a character controller with Three.js is now easier then ever. Checkout the link below for a full tutorial on to build the demo in the video.

🤯 Just build Fortnite in React ... With the new releases of r3/timeline and r3/viverse, building a character controller with Three.js is now easier then ever. Checkout the link below for a full tutorial on to build the demo in the video.

Bela Bohlender

103,714 görüntüleme • 8 ay önce

Jolt Pro is a 100x improvement over existing zkVMs. After months of researching all existing zkVMs, we realized none of them could scale to what we needed with Zero. Then we came across Justin Thaler's work and Jolt, a research project from a16z crypto. We saw massive potential in the math and, finally, a viable way to scale to our needs. So we secretly assembled a team of some of the brightest minds across cryptography, GPU programming, and ASIC design to build an internal Jolt Pro team. Jolt Pro has no precompiles; it runs only RISC-V instructions, without introducing new ad hoc, error-prone constraints. It's impossible to compare against other zkVMs because they are not proving straight RISC-V, most of their work and speed gains exist in dangerous pre-compiles. Jolt Pro scales to infinity. The number of cells you can use in parallel is only limited by the size of the datacenter. Jolt Pro has a path to 4GHz cells using the same configuration we use for our 1.61 GHz cell today. By early 2027, it will be post-quantum and set the standard for all zkVMs. We will eventually make "zero-proofs", and others can try to beat our RISC-V proving in the open. For now, we stay heads down, building Zero to be everything the industry ever wanted. And then just a little more.

Jolt Pro is a 100x improvement over existing zkVMs. After months of researching all existing zkVMs, we realized none of them could scale to what we needed with Zero. Then we came across Justin Thaler's work and Jolt, a research project from a16z crypto. We saw massive potential in the math and, finally, a viable way to scale to our needs. So we secretly assembled a team of some of the brightest minds across cryptography, GPU programming, and ASIC design to build an internal Jolt Pro team. Jolt Pro has no precompiles; it runs only RISC-V instructions, without introducing new ad hoc, error-prone constraints. It's impossible to compare against other zkVMs because they are not proving straight RISC-V, most of their work and speed gains exist in dangerous pre-compiles. Jolt Pro scales to infinity. The number of cells you can use in parallel is only limited by the size of the datacenter. Jolt Pro has a path to 4GHz cells using the same configuration we use for our 1.61 GHz cell today. By early 2027, it will be post-quantum and set the standard for all zkVMs. We will eventually make "zero-proofs", and others can try to beat our RISC-V proving in the open. For now, we stay heads down, building Zero to be everything the industry ever wanted. And then just a little more.

LayerZero

29,409 görüntüleme • 4 ay önce

I vibecoded this neural network visualization for my students and open sourced it. It shows a simple MLP trained on MNIST handwritten digits at several training steps. The visualization is using Three.js and it comes with training code in PyTorch . Link + repo 👇

I vibecoded this neural network visualization for my students and open sourced it. It shows a simple MLP trained on MNIST handwritten digits at several training steps. The visualization is using Three.js and it comes with training code in PyTorch . Link + repo 👇

David Finsterwalder | eu/acc

487,459 görüntüleme • 8 ay önce

A new version of the Ryn and the Thread of Life demo is now live on Steam! We’ve been busy polishing the experience, now is the perfect time to dive back in or jump in for the first time. Try it on Steam today! #indiedev #indiegames #gamedev

A new version of the Ryn and the Thread of Life demo is now live on Steam! We’ve been busy polishing the experience, now is the perfect time to dive back in or jump in for the first time. Try it on Steam today! #indiedev #indiegames #gamedev

Innombrable

24,332 görüntüleme • 2 ay önce

🚨New video today on Line Charts with Victory Native, RN Skia and Expo! The creations of @carloskelly_13 and William Candillon come together beautifully in Victory Native! Demo is running on a low end Samsung A14 so it’s fighting for frames a bit but is still very smooth 📈!

🚨New video today on Line Charts with Victory Native, RN Skia and Expo! The creations of @carloskelly_13 and William Candillon come together beautifully in Victory Native! Demo is running on a low end Samsung A14 so it’s fighting for frames a bit but is still very smooth 📈!

Daniel Friyia 🇨🇦

13,839 görüntüleme • 2 yıl önce

you don’t need a robot to see physical ai in action, you just need a browser. this week I finally tried something I've wanted to do for a while: design an SO-101 scene in three.js, connect it to a real leader arm, collect teleoperation data, train an ACT model, then port it to run in the browser. the result is the first ever demo of ACT that runs locally on the web. try it yourself now at

you don’t need a robot to see physical ai in action, you just need a browser. this week I finally tried something I've wanted to do for a while: design an SO-101 scene in three.js, connect it to a real leader arm, collect teleoperation data, train an ACT model, then port it to run in the browser. the result is the first ever demo of ACT that runs locally on the web. try it yourself now at

Binh Pham

120,373 görüntüleme • 1 ay önce

What you are seeing here is a totally revamped and improved version of a Dreamcast homebrew classic: the Caves tech demo, captured directly from a Sega Dreamcast. FreakDave's original demo already impressed the hell out of me when I first saw it several years ago, getting back into DC development after a several-year hiatus... but this new demo is absolutely INCREDIBLE stuff. It's powered by FreakDave's FDV engine, and now features vertex-based ambient occlusion, light baking, approximated global illumination, fog, and crazy water effects in addition to the dynamic lighting of the original all running at higher FPS! ...and now I can brag about the fact that he's using my SH4ZAM linear algebra library for the DC's SH4 CPU to accelerate all of the math behind the effects! Makes me proud AF! Great work, my dude! Check out the demo here:

What you are seeing here is a totally revamped and improved version of a Dreamcast homebrew classic: the Caves tech demo, captured directly from a Sega Dreamcast. FreakDave's original demo already impressed the hell out of me when I first saw it several years ago, getting back into DC development after a several-year hiatus... but this new demo is absolutely INCREDIBLE stuff. It's powered by FreakDave's FDV engine, and now features vertex-based ambient occlusion, light baking, approximated global illumination, fog, and crazy water effects in addition to the dynamic lighting of the original all running at higher FPS! ...and now I can brag about the fact that he's using my SH4ZAM linear algebra library for the DC's SH4 CPU to accelerate all of the math behind the effects! Makes me proud AF! Great work, my dude! Check out the demo here:

Falco Girgis

15,701 görüntüleme • 4 ay önce

I built a website that builds itself with the new Sonnet and Cursor and open sourced it ...Github link in comment... It took about 2 hours and close to 100 prompts. I actually had added voice mode and proper resizing of components but at some points my single composer which I was using continuously, pooped out when performing checkout and I was very tired in the end and left it like this for now. It is around 1000 lines of code with ~700 lines of JS.

I built a website that builds itself with the new Sonnet and Cursor and open sourced it ...Github link in comment... It took about 2 hours and close to 100 prompts. I actually had added voice mode and proper resizing of components but at some points my single composer which I was using continuously, pooped out when performing checkout and I was very tired in the end and left it like this for now. It is around 1000 lines of code with ~700 lines of JS.

echo.hive

16,815 görüntüleme • 1 yıl önce

Reviewing AI-generated code is the new bottleneck. Writing code was slow before. Now, in a few minutes, you can generate virtually unlimited lines of code using AI. But developers are now spending 70% of their time checking that code. That's the new bottleneck. I've been using Cline, and they have a feature that actually helps with this: You can click a button and have Cline generate an inline explanation for every single line of AI-generated code. Take a look at the attached video. What it does: 1. You can read the code and the explanation together 2. You don't lose context while reviewing code 3. You understand the "why" behind each change And here is the best part: It works on any git diff. You can use it to review PRs, understand commits from other people, or catch up on a codebase you haven't touched in a while.

Reviewing AI-generated code is the new bottleneck. Writing code was slow before. Now, in a few minutes, you can generate virtually unlimited lines of code using AI. But developers are now spending 70% of their time checking that code. That's the new bottleneck. I've been using Cline, and they have a feature that actually helps with this: You can click a button and have Cline generate an inline explanation for every single line of AI-generated code. Take a look at the attached video. What it does: 1. You can read the code and the explanation together 2. You don't lose context while reviewing code 3. You understand the "why" behind each change And here is the best part: It works on any git diff. You can use it to review PRs, understand commits from other people, or catch up on a codebase you haven't touched in a while.

Santiago

64,425 görüntüleme • 7 ay önce

🚀 Excited to release LongLive 2.0! 🎬 An end-to-end infrastructure for long video generation, with FP4 and parallelism at the core of both training and inference. ⚡45.7 FPS generation speed on 5B model⚡ ✨ LongLive 2.0 supports real-video training, few-step distillation, multi-shot training/inference, sequence-parallel acceleration, NVFP4 KV cache, and async VAE decoding deployment. 🧩 To our knowledge, this is the first open-source 4-bit long video generation infra that covers both training and inference. 🙌 Welcome to check it out, try it, and share feedback! 🔗 Code: 📰 Paper: 🎥 Demo: #LongVideoGeneration #VideoGeneration #Realtime #AIInfra #EfficientAI #FP4 #Parallel #NVIDIA

🚀 Excited to release LongLive 2.0! 🎬 An end-to-end infrastructure for long video generation, with FP4 and parallelism at the core of both training and inference. ⚡45.7 FPS generation speed on 5B model⚡ ✨ LongLive 2.0 supports real-video training, few-step distillation, multi-shot training/inference, sequence-parallel acceleration, NVFP4 KV cache, and async VAE decoding deployment. 🧩 To our knowledge, this is the first open-source 4-bit long video generation infra that covers both training and inference. 🙌 Welcome to check it out, try it, and share feedback! 🔗 Code: 📰 Paper: 🎥 Demo: #LongVideoGeneration #VideoGeneration #Realtime #AIInfra #EfficientAI #FP4 #Parallel #NVIDIA

Yukang Chen

59,030 görüntüleme • 2 ay önce

In my previous post about the #Nioh3 demo, I mentioned that when the game isn’t consistently holding 60 FPS or 120 FPS, several issues can show up. Here’s a clearer breakdown of what's going on. The main reason is that the game speed and physics are tied to the frame rate. When you enable the 120 FPS cap, the engine start overreacting and pushes CPU usage very high, even if the game is still running at around 60 FPS (as shown in the first example). In the second clip, you can see the uneven, jittery camera movement I mentioned before. Notice how much smoother 60 FPS and 120 FPS look compared to the unlocked side. Also, camera movement at 60 FPS is slightly faster than at 120 FPS. In the third clip, player movement is slower when the frame rate is unlocked. This doesn’t happen all the time, but it shows up often enough to be noticeable. Because of how the Katana Engine behaves, the game is clearly designed around 60 FPS. Running at 120 FPS is possible, but it’s only recommended if your system can maintain that target almost all the time, which isn’t easy to achieve. There’s also an alternative workaround where you select the 60 (locked) option and enable Frame Generation (DLSS or FSR 3), as shown in the last clip. The downside is that DLSS Frame Generation tends to show the same stuttery look as when the frame rate isn’t holding a fixed target, likely due to Reflex keeping the frame rate slightly below target. FSR Frame Generation, on the other hand, looks much smoother and works better here.

In my previous post about the #Nioh3 demo, I mentioned that when the game isn’t consistently holding 60 FPS or 120 FPS, several issues can show up. Here’s a clearer breakdown of what's going on. The main reason is that the game speed and physics are tied to the frame rate. When you enable the 120 FPS cap, the engine start overreacting and pushes CPU usage very high, even if the game is still running at around 60 FPS (as shown in the first example). In the second clip, you can see the uneven, jittery camera movement I mentioned before. Notice how much smoother 60 FPS and 120 FPS look compared to the unlocked side. Also, camera movement at 60 FPS is slightly faster than at 120 FPS. In the third clip, player movement is slower when the frame rate is unlocked. This doesn’t happen all the time, but it shows up often enough to be noticeable. Because of how the Katana Engine behaves, the game is clearly designed around 60 FPS. Running at 120 FPS is possible, but it’s only recommended if your system can maintain that target almost all the time, which isn’t easy to achieve. There’s also an alternative workaround where you select the 60 (locked) option and enable Frame Generation (DLSS or FSR 3), as shown in the last clip. The downside is that DLSS Frame Generation tends to show the same stuttery look as when the frame rate isn’t holding a fixed target, likely due to Reflex keeping the frame rate slightly below target. FSR Frame Generation, on the other hand, looks much smoother and works better here.

BenchmarKing

17,820 görüntüleme • 6 ay önce

GTA-Style Open World Playground, Built Entirely Through Prompts with ALCHEMIST AI 🔮. This one was generated entirely in-browser using our Three.js engine, with support for external 3D models, custom physics, vehicle logic, and interactive elements. Want to create a superhero city game? A heist getaway sim? A surreal low-poly driving world with AI NPCs? Just describe it. Alchemist AI handles the rest. And with v3.5 on the horizon, the possibilities are about to multiply.

GTA-Style Open World Playground, Built Entirely Through Prompts with ALCHEMIST AI 🔮. This one was generated entirely in-browser using our Three.js engine, with support for external 3D models, custom physics, vehicle logic, and interactive elements. Want to create a superhero city game? A heist getaway sim? A surreal low-poly driving world with AI NPCs? Just describe it. Alchemist AI handles the rest. And with v3.5 on the horizon, the possibilities are about to multiply.

ALCHEMIST AI 🔮

13,854 görüntüleme • 1 yıl önce

Infinity IDE has officially integrated the x402 protocol — marking the arrival of Pay-Per-Use Payments × Vibe Coding. Now you can integrate x402 payments into your apps using natural language—unlocking flexible monetization for your creations. Multi-chain support is live, starting with BNB Chain and Base, making it easier than ever to build, ship, and monetize on-chain. Watch our tutorial and try it yourself in the #AgenticIDE:

Infinity IDE has officially integrated the x402 protocol — marking the arrival of Pay-Per-Use Payments × Vibe Coding. Now you can integrate x402 payments into your apps using natural language—unlocking flexible monetization for your creations. Multi-chain support is live, starting with BNB Chain and Base, making it easier than ever to build, ship, and monetize on-chain. Watch our tutorial and try it yourself in the #AgenticIDE:

InfinityGround

34,967 görüntüleme • 7 ay önce

MOST PEOPLE GO THROUGH LIFE CHASING THEIR TAILS HALF ASLEEP 😴 YOU WATCH AND LEARN HOW TO HELP YOURSELF WITH PLANT 🌱 Supreme CBD - UK’s #1 CBD OIL contains 16 cannabinoids which work together to heal our bodies from within using our bodies OWN ENDO CANNABINOID SYSTEM 🌱 these are all facts and can’t be denied, if you want to try get a 6000MG oil WITH FREE £55 4800MG CBD GUMMIES with code FOWLER40 on the link in my bio 🙏🏻 #supremecbd #ad.

MOST PEOPLE GO THROUGH LIFE CHASING THEIR TAILS HALF ASLEEP 😴 YOU WATCH AND LEARN HOW TO HELP YOURSELF WITH PLANT 🌱 Supreme CBD - UK’s #1 CBD OIL contains 16 cannabinoids which work together to heal our bodies from within using our bodies OWN ENDO CANNABINOID SYSTEM 🌱 these are all facts and can’t be denied, if you want to try get a 6000MG oil WITH FREE £55 4800MG CBD GUMMIES with code FOWLER40 on the link in my bio 🙏🏻 #supremecbd #ad.

Anthony Fowler

83,947 görüntüleme • 9 gün önce

Today, we're introducing the Consensus Framework to the healthcare industry! We’re a 20-person YC startup that designed a framework surpassing OpenAI & Paul Jankura in healthcare benchmarks. Our new decision-making AI just hit a record-breaking 60% on MedExpertQA benchmarks (all-time high). Try it out yourself here: You can also sign up for the API waiting list; link in the comments.

Today, we're introducing the Consensus Framework to the healthcare industry! We’re a 20-person YC startup that designed a framework surpassing OpenAI & Paul Jankura in healthcare benchmarks. Our new decision-making AI just hit a record-breaking 60% on MedExpertQA benchmarks (all-time high). Try it out yourself here: You can also sign up for the API waiting list; link in the comments.

Ahmed Omar.

26,048 görüntüleme • 1 yıl önce

Take any Python code and build a dynamic GUI on top of it by dragging and dropping components. Python is already the most popular language in the world. This makes it even better! The attached video is only 45 seconds. It will show you how it works. Here is the important part: You can now build an interface for your Python code without learning HTML, JavaScript, or CSS. Go to this link to try this: Remember to work smarter, not harder! Thanks to the team behind Taipy for partnering with me on this post.

Take any Python code and build a dynamic GUI on top of it by dragging and dropping components. Python is already the most popular language in the world. This makes it even better! The attached video is only 45 seconds. It will show you how it works. Here is the important part: You can now build an interface for your Python code without learning HTML, JavaScript, or CSS. Go to this link to try this: Remember to work smarter, not harder! Thanks to the team behind Taipy for partnering with me on this post.

Santiago

155,179 görüntüleme • 2 yıl önce

We just shipped a massive upgrade to our design agent.🚢 We’ve tuned it to push Gemini 2.5 Pro to its maximum creative limits. The results? More interesting layouts, richer colors, sharper copy, and better font/photo pairings. In our testing, users preferred the new results by a significant margin. See the comparison for yourself in the video (old is on the left, new is on the right) [sound on!]. Try it on something new or take an existing project and re-run your prompt.

We just shipped a massive upgrade to our design agent.🚢 We’ve tuned it to push Gemini 2.5 Pro to its maximum creative limits. The results? More interesting layouts, richer colors, sharper copy, and better font/photo pairings. In our testing, users preferred the new results by a significant margin. See the comparison for yourself in the video (old is on the left, new is on the right) [sound on!]. Try it on something new or take an existing project and re-run your prompt.

Stitch by Google

136,122 görüntüleme • 9 ay önce