Zephyrus G16 Hits 215 W in Experimental Power Limit Test
An ASUS ROG Zephyrus G16 equipped with an RTX 5090 Laptop GPU has been pushed to roughly 215 W of combined CPU and GPU power using an experimental NVIDIA power control utility. The test shows that thin gaming laptops can operate beyond their factory power targets, but temperatures approaching thermal limits and fans running close to maximum speed also demonstrate why manufacturers impose those limits in the first place.
The experiment used NvpwrControl, an unofficial tool capable of modifying runtime power targets on several GeForce RTX 50 Series Laptop GPUs without flashing the VBIOS. Its developer describes support for experimental RTX 5080 and RTX 5090 configurations with selectable GPU targets reaching as high as 225 W, although actual behavior depends heavily on the individual laptop and its cooling and power delivery design.
Testing shared by Reddit user Firm Age4730 showed the RTX 5090 reaching around 185 W while the Intel Core Ultra 9 386H consumed approximately 31 W during Cyberpunk 2077, putting combined CPU and GPU consumption slightly above 215 W. This is substantially more aggressive than the factory configuration of the 2026 Zephyrus G16, for which ASUS officially lists a maximum RTX 5090 TGP of 160 W in Manual Mode.
The additional power does address one limitation observed by the user. Under demanding gaming loads at the factory configuration, the GPU could consume most of the available thermal and electrical budget, leaving the processor operating at around 13 W. Increasing the available power allowed both components to consume more simultaneously, but the cooling system quickly became the next constraint.
During testing, the RTX 5090 reached 84°C while the Core Ultra 9 386H climbed to 95°C. Those figures place both components close to their thermal operating limits. The cooling fans were also running at approximately 6700 RPM for the CPU and 6800 RPM for the GPU, making sustained operation considerably louder than the factory configuration.
That tradeoff is particularly relevant to the Zephyrus G16 because ASUS has built the system around portability as much as performance. The 16 inch laptop measures only 1.49 cm thick and weighs around 1.85 kg, using a vapor chamber, 3 fan cooling system, liquid metal, and revised exhaust design to manage its hardware within a relatively compact chassis.
Larger gaming laptops can dedicate considerably more physical space to cooling and power delivery. We have previously covered RTX 5090 gaming laptops using much more aggressive power configurations, including systems designed around larger cooling assemblies and even external liquid cooling. Thin systems such as the Zephyrus G16 are balancing a different set of priorities.
There is also an important distinction between demonstrating that a laptop can briefly operate at these power levels and proving that it should. NvpwrControl itself is described as experimental, requires Windows Driver Signature Enforcement to be disabled in its current form, and warns that increased power targets can raise GPU, memory, cooling system, and power delivery stress.
This experiment is less about finding free performance and more about showing where thin gaming laptop design reaches its physical limits.
The Zephyrus G16 can clearly feed substantially more power to its RTX 5090 and processor than ASUS allows under its normal configuration. The problem is removing one restriction immediately exposes another. Once power is increased, cooling capacity, temperature, fan speed, and power delivery become the limiting factors.
That does not mean ASUS is leaving performance unnecessarily locked away. A 1.49 cm chassis has to balance performance against noise, surface temperature, component longevity, and everyday usability. Pushing more than 215 W through the CPU and GPU may be technically possible, but sustaining it comfortably is a very different challenge.
For users who prioritize maximum performance above portability, thicker gaming laptops remain better suited to high sustained power. For machines built around mobility, the factory power limit is part of the engineering compromise rather than simply an artificial restriction.
Would you accept higher temperatures and much louder fans for extra laptop performance, or would you rather keep a thinner gaming system within its factory power limits?
