NVIDIA GeForce RTX 4090 Laptop GPU Pulls up to 200W, GA103-based, Lineup Power Detailed

At its 2023 International CES event, NVIDIA is expected to launch not just its desktop GeForce RTX 4070 and RTX 4070 Ti graphics cards, but more importantly, also its GeForce RTX 40-series Laptop GPU series powering next-generation gaming notebooks based on the upcoming 13th Gen Core “Raptor Lake” processors. NVIDIA seems to be making a very tight rope-walk between power-management and generational performance increase in this power- and thermal-constrained form-factor. Wccftech scored a major scoop on the specs of various RTX 40-series Laptop GPUs.

The GeForce RTX 40-series “Ada” Laptop GPU lineup will be led by the RTX 4090 Laptop GPU, based on the 4 nm “AD103” silicon (same one that powers the desktop RTX 4080). It will be equipped with 16 GB of memory, a yet-unknown core-configuration, GPU Boost frequencies of up to 2.04 GHz, and typical power draw ranging between 150 W to 175 W, which can peak up to 200 W thanks to the 25 W dynamic boost range (power permissible by the platform if the other components such as CPU aren’t drawing their peak power).

Moving down the lineup, we see the RTX 4080 Laptop GPU being based on the “AD104,” the same silicon that powers the upcoming desktop RTX 4070 Ti and RTX 4070. The core-configurations remain unknown, but given the naming, the RTX 4080 Laptop GPU likely maxes it out. The SKU features 12 GB of memory across the chip’s 192-bit memory bus, boosts up to 2.28 GHz, and has a similar power-limit to the RTX 4090 Laptop GPU—150 W to 175 W typical graphics power, with additional room for 25 W.

The RTX 4070 Laptop GPU is based on the yet-unreleased 4 nm “AD106” silicon which, unless we’re mistaken, features a 128-bit GDDR6X memory interface, giving this SKU its 8 GB of memory. This GPU gets typical graphics power in the range of 115 W to 140 W, with room for 25 W more. Moving lower down the stack, we have the RTX 4060 Laptop GPU, and the RTX 4050 Laptop GPU, both of which are based on the 4 nm “AD107” silicon, which is probably the smallest “Ada” implementation. The former gets 8 GB of memory, and the latter 6 GB of it.