Volume ray marching
Trace rays through a mathematical volume; control detail and sampling steps.
Volume Ray Marching Stress Test BROWSER GRAPHICS & COMPUTE
GPU Video Texture Stress Test.
Render a locally selected video as a browser GPU texture and inspect delivered frames and frame timing. Codec support and hardware decoding vary.
WEBGL2960 × 540
ANGLE (NVIDIA, NVIDIA GeForce RTX 4050 Laptop GPU (0x000028A1) Direct3D11 vs_5_0 ps_5_0, D3D11)Local video delivery—FPS
Frames/s— FPS
Frame interval— ms
GPU timing— ms
Elapsed0 / 15 s
Frames/s1 FPS
Start light. Stop if heat, fan noise, or responsiveness becomes uncomfortable. This page cannot read temperature, power, or GPU utilization.
Run environmentWEBGL2
- Renderer
- ANGLE (NVIDIA, NVIDIA GeForce RTX 4050 Laptop GPU (0x000028A1) Direct3D11 vs_5_0 ps_5_0, D3D11)
- Reported vendor
- Google Inc. (NVIDIA)
- Browser
- Chrome 155.0.0.0
- Operating system
- Windows
- Actual API
- webgl2
- Fallback reason
- Unavailable
- Timing method
- gpu-timer-query
- Workload version
- ioc-gpu-lab-v1
02 / Summary
Run results
Run a test to see measured results.
03 / Runs stay in this browser until you choose to export, share, or publish.
Local history
Clearing local history does not withdraw published results. Private removal credentials stay in this browser separately; download backups before clearing browser storage.
No saved runs in this browser.
04 / Compare runs
Public results
Public reports include measurement samples, settings, GPU/vendor labels, browser/OS labels, and run/submission dates. Results are self-submitted and not independently verified. Runs are retained for up to 365 days, unless withdrawn sooner.
Withdraw using a private backup
If browser storage was cleared, enter the report ID and deletionToken from your private JSON backup. Never paste the code into a URL or public message.
05 / Measurement methodology
Measurement methodology
Warm-up frames are excluded from the measured summary.
Results describe this browser workload and its recorded settings. Frame timing cannot identify the cause of a slowdown. These tests do not measure temperature, power, or total VRAM and do not certify hardware health. Compare only matching workloads, APIs, settings, and timing methods.
Measurement methodologyShader arithmetic
Repeat pixel shader calculations to vary arithmetic cost.
Fragment Shader Stress TestGeometry throughput
Submit triangles and observe browser rendering throughput.
GPU Geometry Stress TestInstanced objects
Draw repeated objects with shared geometry and per-instance transforms.
GPU Instancing Stress TestParticle field
Render many moving points; vary particle count and pixel load.
GPU Particle Stress TestParticle simulation
Simulate gravity and collisions on the CPU, then draw particles on the GPU. This measures a mixed workload.
GPU Physics Stress TestShader ray tracing
Trace a small procedural scene in a shader; no hardware ray-tracing claim.
GPU Ray Tracing Stress TestPost-processing passes
Apply image effects through repeated render passes.
GPU Post Processing Stress TestLighting and shadows
Increase shader lighting and shadow sampling work.
GPU Shadow Stress TestTexture sampling
Sample generated textures within the selected allocation budget.
GPU Texture Bandwidth Stress TestTexture round-trip check
Write known texture patterns and compare sampled readback values. This does not test all physical VRAM.
GPU Texture Integrity TestWebGPU compute
Run a WebGPU compute kernel and inspect its returned values and timing.
WebGPU Compute Stress TestLocal video delivery
Observe delivery of frames from a video you select on this device.
GPU Video Texture Stress TestTexture upload
Transfer generated pixel data into textures during the run.
GPU Texture Upload Stress TestCustom shader
Edit a local shader experiment. Results are separate from standard benchmarks.
Custom GPU Shader Stress Test