Batched Rendering

Rendering camera images for hundreds or thousands of parallel environments in one pass, feeding vision policies at high speed.

GPU-parallel simulation can run thousands of environments at once, but if each environment's camera were rendered and copied separately, image bandwidth would become the bottleneck: Isaac Lab's documentation estimates that one 800×600 floating-point image is nearly 2 MB, so 60 frames per second alone needs 120 MB/s, before multiplying by the number of cameras and environments. Batched rendering merges the processing of the same camera across all environments. Isaac Lab's TiledCamera (supported from Isaac Sim 4.2.0) is a good example: all the cloned cameras share a single render product, and each environment's image is tiled into one large combined image that can be handed to the training code in a single sync. ManiSkill3's SAPIEN-based parallel rendering and MuJoCo Playground's built-in batched renderer offer similar capability. This is what lets pixel-based vision-motor policies be trained with massively parallel reinforcement learning too.

ExampleIn Isaac Lab, setting a TiledCameraCfg's camera path to /World/envs/env_.*/Camera retrieves an 80×80 RGB image for every environment in one call; the official guidance suggests around 512 cameras on an RTX 4090-class GPU.

Also called

Tiled Rendering, Tiled Camera, Batch Renderer

Related

GPU-Accelerated Parallel Simulation · Vectorized Environments · Rendering · NVIDIA Isaac Lab · ManiSkill · Massively Parallel Reinforcement Learning

Sources

Isaac Lab Documentation: Camera (Tiled Rendering)
ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI (arXiv 2410.00425)
MuJoCo Playground (arXiv 2502.08844)

As of

2026-09