GeForce RTX 4070 SUPER, RTX 4070 Ti SUPER, and RTX 4080 SUPER are ready from NVIDIA.

The higher end of NVIDIA’s GeForce RTX 40-series “Ada” series is said to be getting an update, according to hongxing2020, a dependable source with NVIDIA leaks. It’s possible that the business may revive the SUPER brand extension it first established with the RTX 20-series. GeForce RTX 4080 SUPER, GeForce RTX 4070 Ti SUPER, and GeForce RTX 4070 SUPER are the potential three SKUs.

There is no information available regarding the company’s release date or the details of them, but we can undoubtedly make assumptions. While the current RTX 4080 is based on the AD103 silicon, it does not utilize all 80 of the SM (streaming multiprocessors) available on the chip. Given that the 256-bit memory bus of the AD103 is at its maximum capacity, we doubt that the additional 4 SM would be able to increase performance enough to create a completely new SKU. We forecast that the RTX 4080 SUPER may be built on the larger AD102 silicon, which really has 144 SM, of which the RTX 4090 currently uses 128. NVIDIA offers the option of selecting an SM count, like, say, 96. The RTX 4080 SUPER may receive the same 24 GB of memory as the RTX 4090 thanks to AD102’s bigger 384-bit memory bus, or even 20 GB via a 320-bit memory bus.

As for the RTX 4070 Ti SUPER (an curiously titled SKU that contains both the Ti and SUPER brand extensions), it stands to reason that this new SKU could be based on AD103, with a lower SM count than the 76 of the RTX 4080, given that the existing RTX 4070 Ti maxes out the AD104 silicon (60 SM). It is unknown whether NVIDIA will use the entire 256-bit memory bus. Regarding the RTX 4070 SUPER, the difference between the current RTX 4070 Ti and the RTX 4070 is quite large. The RTX 4070 only uses 46 SM on the AD104, whereas the RTX 4070 Ti utilizes all 60. This allows NVIDIA the chance to strengthen its position against AMD’s Radeon RX 7800 XT and draw closer to the RTX 4070 Ti in terms of SM counts.