Kopite7kimi reports that NVIDIA’s GeForce RTX 50 “Blackwell” GPU family will commence with the RTX 5080 graphics card, followed by the flagship RTX 5090. As for the long-awaited launch time of the NVIDIA GeForce RTX 50 “Blackwell” GPU family, we expect it to hit the market in Q4 2024. However, there are rumors that it will start with two products: the GeForce RTX 5090 and the GeForce RTX 5080.
More About the NVIDIA GeForce RTX 5080
It’s important to note that unlike the “Ada Lovelace” lineup, where NVIDIA rolled out the GeForce RTX 4090 first and then the RTX 4080, this time NVIDIA appears to have launched the “80” model first. It is worth saying that this leaks shortly after a few days, and a few media outlets implied that NVIDIA only intends to release the GeForce RTX 5090 model alone this year. However, this is not right as previous leaks have confirmed that both RTX 5090 and RTX 5080 will be eligible for launch.
In this case, Kopite7kimi has backed those claims mentioned earlier. Nevertheless, NVIDIA has not yet hinted or presented any details about its next-generation GPU family. The known NVIDIA Blackwell GPUs have been planned for artificial intelligence and high-performance computing support, as there is a known NVIDIA Blackwell GPU IP to drive the next-generation RTX graphics models.
The original rumors suggested that NVIDIA would have two flagship dies, codenamed GB202 and GB203. The first one will be used by the GeForce RTX 5090 and feature up to 192 SM units and a 512-bit interface, while the second one will power the RTX 5080. It is rumoured that the GB203 will be half the size of the GB202, with 92 SM units and a 256-bit interface.
Both chips will use a novel GDDR7 memory interface. It is hard to believe that announcements regarding the gaming lineup during Computex 2024, but whispers from the AIB partners could start to appear soon. In the best-case scenario, the first teaser campaign might start by late Q3.
