System information
These values are browser-provided hints, not a hardware inventory. The memory hint is neither installed RAM nor free memory: a device with 32 GiB installed can report a smaller value. Vigilum cannot determine your actual capacity from this hint; check your operating system for installed RAM. Privacy protections can also limit processor counts and mask graphics names. Compare runs in the same browser and configuration.
CPU benchmark
Measure CPU render speed. Choose a workload, then start.
Start a render to build the image.
Worker output
Output activity, not physical core utilization. No workers are currently marked active.
The image builds as workers finish. Display updates are separate from measured render time.
4 browser-reported threads · automatic workers
Save other work. This test can slow your device.
Ready
Measurements
- Measured time
- —
- Average primary rays/s
- —
- Visible tiles
- —
- Sanity errors
- —
Tracing throughput
Cumulative average rate samples · last 0 observations · M rays/s
A throughput trend appears when measurements arrive.
Method & safety
Selected workload: 1920 × 1080 · 192 samples/pixel · Fixed work; no time limit. Settings are locked while running. Switching modules stops the current workload.
Workers render one deterministic image. Points equal 1,000 divided by render time in seconds. Timing starts after worker warm-up and ends when the last worker finishes; display painting is independent of computation. Continuous mode repeats work without timer sleeps. Cancellation terminates workers. Browser scheduling may prevent full system utilization. Zero sanity errors does not certify hardware health.
The worker traces a fixed sphere scene using a bounding-volume hierarchy to skip unrelated objects. Standard and continuous runs use 192 samples per pixel; Extreme uses 1,920. The maximum path depth is 5 and output arrives in 16 × 16 pixel tiles (edge tiles may be smaller). Samples are deterministic; repeated sweeps render the same scene, not an animation.
Primary rays/s = total primary samples across workers / elapsed seconds. Secondary path bounces are not additional primary rays. Finite elapsed time is the maximum reported worker time from a shared start epoch, excluding warm-up but including scheduling delays. The live finite rate is cumulative; continuous rate uses recent worker heartbeats. The points formula is an app-defined inverse-time index, not a calibrated processor rating. A worker-normalized rate is not a separately measured single-core score.
Continuous display clears on the first tile of a newer dispatched sweep. Older-sweep output is then ignored for display only, even if a slower worker is still finishing it. The sweep label describes displayed output, not a global completion barrier. Stopping terminates workers; partial output never becomes a finite completed score.