Nvidia RTX 4090 is up to 4x faster than 3090 Ti but takes 450W of power to do so

As expected, Nvidia has launched its highly anticipated RTX 4000 series lineup of graphics cards at GTC 2022 today. The new GPUs are based on the Ada Lovelace architecture which brings significant improvements to the core architecture delivering up to 90 Shader TFLOPs, twice the power efficiency, 200 RT TFLOPs, and much more. The improvements are highlighted in the image below:

RTX 4000 series is based on the TSMC 4N (4nm) process and despite the improvements to efficiency, Nvidia has had to push the cards to 450W in order to reach these levels of performance. The chart below shows the improvement in raw throughput on Ada Lovelace and compares it against the power draw increase. For those that desire the best, cutting-edge performance out there, the extra power draw on the 40-series cards will probably be worth it.

The RTX 4090, according to Nvidia, is anywhere between two times to four times faster compared to the RTX 3090 Ti. The card comes with 24GB GDDR6X VRAM and is priced at $1,599, which is $100 more than the RTX 3090 at launch.

Up next, we have the RTX 4080 which comes in two flavors, the RTX 4080 16GB and the RTX 4080 12GB. These succeed the RTX 3080 12GB and RTX 3080 10GB respectively.

The two models are priced at $1,199 and $899 respectively and will be available in November, which is also when AMD will unveil its next gen RDNA 3 Radeon RX 7000 series GPUs. Like with the RTX 4090, Nvidia claims the RTX 4080 is also two to four times faster than the RTX 3080 Ti.

Nvidia RTX 4090 is up to 4x faster than 3090 Ti but takes 450W of power to do so