NVIDIA has uploaded its whitepaper for the RTX “Blackwell” GPU architecture, and with it comes 57 pages of insight surrounding the GPUs that power the new GeForce RTX 50 Series, including the flagship GeForce RTX 5090 and GeForce RTX 5080, both of which are on track to release tomorrow (Jan. 30) for $1,999 and $999, respectively.
“The first GeForce RTX 50 series product that will be launching using the GB203 GPU is the GeForce RTX 5080,” NVIDIA confirms in a portion of the whitepaper before going on to list the specifications for the GeForce RTX 5080 and RTX 5070 Ti, the latter of which leverages the very same GPU.
“The Ada GB205 GPU is the perfect entry point for gamers, content creators, and streamers who want the new features NVIDIA is introducing with the Blackwell GPU architecture,” the company goes on to write before sharing the specs of the GeForce RTX 5070, which additionally counts 12 GB of GDDR7 with a memory bandwidth of 672 GB/s among its hardware.
“The NVIDIA RTX Blackwell architecture features AI multi-frame generation that boosts DLSS 4’s frame rates up to 2x over the previous DLSS 3/3.5, while maintaining or exceeding native image quality and providing low system latency,” reads a portion of the whitepaper breaking down all of the key features enabled by green team’s latest tech, echoing something that the company had mentioned previously during the unveil of the GeForce RTX 50 Series in early January.
“With the invention of DLSS 4 with Multi Frame Generation, working in unison with the complete suite of DLSS technologies, we can multiply frame rates by up to 8X over traditional brute-force rendering and provide image quality that is better than native rendering,” NVIDIA’s Andrew Burnes noted on his post about how Blackwell would deliver game-changing AI and neural rendering capabilities to gamers and creators.