NVIDIA’s Ampere GPU architecture, along with its corresponding GeForce RTX 30 Series graphics cards, have been on the market for nearly a year now. To that end, it comes as no surprise that rumours and speculations about the GPU maker’s next graphics architecture have begun cropping up online.
The latest rumours pertaining to NVIDIA’s after-Ampere GPU, is that it could go by the name Ada Lovelace and that its GPU design has been somewhat finalised, according to a leakster who goes by the handle Greymon55 (@greymon55). In addition to this detail, they also say that the Ada Lovelace will be based on TSMC’s 5nm process node.
As exciting as this sounds, prominent leakster kopite7kimi (@kopite7kimi), who initially predicted the entire RTX 30 Series lineup, isn’t quite as sure that NVIDIA will go through with Ada Lovelace as the GPU architecture for its so-called RTX 40 Series graphics cards. To be precise, kopite7kimi says that NVIDIA’s decision may end up hinging on the success of AMD’s own next-generation RDNA3 architecture, which is expected to utilise its multi-chip mode (MCM) design.
In addition, YouTuber Moore’s Law is Dead also suggested that Ada Lovelace may not just be limited to desktop GPUs and could also end up being a primer for future NVIDIA SoCs. The YouTuber says that the GPU may just end up being a part of NVIDIA’s Orin Series SoCs that are used in its self-driving card initiative.
-5nm(no matter TSMC or SEC)
-AD102 in transition or GH202 in revolution?
-RDNA3, GFX11, how does AMD reach its perf goal? Double the SIMD in CU?
-How about Intel?
— kopite7kimi (@kopite7kimi) May 22, 2021
In any case, the one speculation that does seem to bring all the leakster to agreement is Ada Lovelace’s potential launch date: at the very latest, they believe that a graphics card equipped with the GPU may only appear sometime nearing the end of 2022, which is more or less in line with NVIDIA’s current GPU cycles. Fingers crossed, the brand will at least have something ready for the tech media by September next year.