AMD launched the Navi GPUs based on the new RDNA architecture at E3 promising improved performance over the GeForce RTX GPUs all the while touting a significantly improved power draw as well thanks to the new architecture and TSMC’s 7nm process. However, right after the event, NVIDIA claimed that the 12nm based Turing GPUs are still the more power efficient GPUs, thanks to the company’s superior design.

AMD’s Navi parts are fabbed on the 7nm node while the NVIDIA Turing chips use the older 12nm process. According to AMD, the newer process decreased the TDP by 30% while the rest came from RDNA’s architectural enhancements. Despite that NVIDIA appears adamant that the Turing architecture powers “the most power efficient GPUs on the planet.”

Turing is a 12nm part. And we’re quite proud of that. Turing remains the most power efficient GPU on the planet. Over Pascal, it is up to 1.4x our efficiency. If I had to draw on here. We don’t like to revel in our competition. But I will say there’s another GPU that was talked about that is roughly 2070 performance. The power of that device is right about here. If I had to draw where the Radeon VII might be it’s roughly 2080 performance. And it’s somewhere out here for power.

Jeff Fisher, SVP of GeForce

According to the official specs, yes the NVIDIA RTX GPUs despite having the additional RTCores and Tensors for ray-tracing and DLSS are more power efficient compared to AMD Navi, but like all first-party figures you ought to take these with a grain of salt. The RX 5700 GPUs are still not available to 3rd party reviewers. It’s only after rigorous testing that this claim can be verified.

Read more:

Leave a Reply