Bo Peng (@Builtin_Pb) profile picture

CS PhD @ Princeton

Joined October 2024
Photo by Bo Peng (@Builtin_Pb)Photo by Bo Peng (@Builtin_Pb)Photo by Bo Peng (@Builtin_Pb)
robotics is accelerating physics + rendering + training 3 in 1 when
Dear researchers: Did you know that your H100 can actually ray-trace like a gaming card? 🎮 for a long time, people have assumed that compute cards like the H100 can only compute, and that you need a graphics card like the RTX 5090 to render. we show that's wrong. 🚨Meet RTX H100. Blender Cycles on the Open Data benchmark (Blender 5.3.0): ours 48,641. the RTX 5090 is at 19,202, and a stock H100 gets only 7,041. our original goal was to match the 5090. we ended up at 2.5x. our neural rendering approach is lossy. we test within a reasonable loss, and every result passes an image check. we think nobody doing this before was an accident: graphics people rarely touch compute cards, and people who have compute cards rarely know graphics. robotic simulators need fast rendering. now a general-purpose compute card can do it, so you don't need heterogeneous compute or a complicated cluster. we'll extend this to more hardware next. almost ready for GTA 6 on the cluster 🤣 Best, Wenhao
1
1
9
508
What does RoboDojo really simulate? We followed it from the real world to its score, one link at a time. At almost every step, it simulates something other than what it says. Same bottle, same grasp. As shipped, it tips 78° in the hand. After our fix, it stays level.
7
18
15
170
36,864
4/ Values no real thing has: five of the arm’s six joints may turn 573° each way. The real arm stops within half a turn. About half the benchmark’s released demonstrations bend a joint further than the real arm can. The garbage is already in the dataset.
1
6
1,426
FrontierCS 2: A Benchmark for Computer Science R&D
28
26
12
379
42,068