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.
Combined burn-in
Concurrent CPU tracing, graphics, and memory verification. Measurement starts after all subsystems are ready.
16 raymarched spheres · 32 primary march steps. Fixed render size, independent of panel size and display scaling. The initial settings are candidates, not hardware-calibrated defaults. All profiles can affect responsiveness.
Custom session settings
CPU tracing and RAM verification stay continuous. GPU settings are fixed for each run; there is no automatic load ramp or CPU-renderer fallback. All three subsystems run together.
Ready
Live measurements
- Measured duration
- 0 s
- CPU recent rate
- —
- GPU recent rate
- —
- RAM recent rate
- —
- Validation errors
- —
Concurrent workload
Combined workload
Three subsystems. One session.
- CPU
- Continuous path tracing
- GPU
- Spheres · 960 × 540
- RAM
- Repeated pattern verification
Choose duration and memory budget, then start. Separate CPU and GPU viewports show their workloads; RAM verification measurements appear alongside them.
Save other work before starting. Stop releases every subsystem; preparation times out after two minutes.
Measurements & method
Combined method & calculations
The coordinator runs continuous CPU path tracing, the selected Three.js/WebGL2 GPU profile and continuous RAM verification together. GPU render dimensions are fixed at pixel ratio 1; resizing the panel does not change the workload. Settings are recorded for each run. Compare only matching profiles, resolutions and workload versions. Legacy Combined results used a canvas-sized Heavy shader and are not directly comparable to these profiles.
After all subsystems are ready, a one-second tick records index = round((CPU primary rays/s ÷ 1,000,000) × 10 + GPU FPS × 5 + RAM GiB/s × 20). These app-defined weights combine unlike units and are not a normalized benchmark score. Sustained throughput is the arithmetic mean of sampled indices. The baseline is the largest index sampled in the first five measured seconds.
Performance retained = min(100, round(mean index / max(1, baseline) × 100)). It describes relative browser-workload throughput, not measured thermal throttling. Scheduling, background activity and browser limits can change it. No temperature, power or hardware utilization is measured. Stop releases every subsystem; preparation has a two-minute timeout.
The composite index combines unlike workloads with fixed weights. It is not temperature, hardware utilization, a CPU render-speed score, or a certification of system health.