GPU Power Draw Simulator: Frame Rate vs Energy Efficiency

Render Settings

Resolution

Graphics quality

Frame-rate cap 144 fps

GPU power limit 100%

Live Readouts

Power draw

— W

Achieved frame rate

— fps

Efficiency

— fps/W

Energy per hour

— Wh

Cool & efficient

How it works

A GPU's board power is dominated by dynamic power dissipation, which scales super-linearly with clock frequency because raising frequency also requires raising core voltage (dynamic voltage-frequency scaling, DVFS). This sim models it as a power law:

P = P_idle + a · f^n          (n ≈ 2.3, the combined V(f)·f scaling)
workload W = quality_load × resolution_factor
required clock f = FPS × W × k
FPS_achieved = min(FPS_cap, FPS_max(power_limit))
efficiency = FPS_achieved / P   (frames per watt)

Resolution / quality — more pixels and heavier shading (shadows, ray tracing) raise the workload W, so the same frame rate needs a higher clock — and power grows as roughly the square of that clock, not linearly.
Power limit — caps the maximum sustainable clock the card is allowed to reach, which caps the maximum achievable frame rate at a given workload.
Frame-rate cap — this is the real energy-saving lever: once FPS is capped below the card's ceiling, the GPU idles down its clock between frames, so power drops roughly with the square of the frame-rate reduction. Capping 240 fps down to 60 fps at the same settings can cut power draw several times over.

This is the same trade-off real GPU vendors expose as "frame rate limiter" and "power limit / TDP slider" features, and the one streaming/cloud-gaming and data-center operators tune at scale to cut electricity cost and cooling load.

Heatsink temperature Cooling fan

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.