Tuesday, October 17th 2023
NVIDIA Readies GeForce RTX 4070 SUPER, RTX 4070 Ti SUPER, and RTX 4080 SUPER
NVIDIA is rumored to be working on a refresh of the higher end of its GeForce RTX 40-series "Ada" series, according to hongxing2020, a reliable source with NVIDIA leaks. The company could be bringing back the SUPER brand extension that it introduced with the RTX 20-series. As many as three SKUs are on the radar—GeForce RTX 4080 SUPER, GeForce RTX 4070 Ti SUPER, and the GeForce RTX 4070 SUPER.
There is no word on when the company plans to release these, or what their specifications are, but we can certainly speculate. The current RTX 4080, while based on the AD103 silicon, doesn't max it out—it uses 76 out of 80 SM (streaming multiprocessors) available on the silicon, but we doubt if those extra 4 SM could drive up enough performance to make a whole new SKU, especially given that the 256-bit memory bus of the AD103 is maxed out. We predict that the RTX 4080 SUPER could be based on the larger AD102 silicon that physically has 144 SM that the current RTX 4090 uses 128 out of. NVIDIA has the opportunity to pick an SM count such as, say, 96. AD102 also has a wider 384-bit memory bus, giving NVIDIA the option of either giving the RTX 4080 SUPER the same 24 GB memory configuration as the RTX 4090, or even 20 GB, across a 320-bit memory bus.
As for the RTX 4070 Ti SUPER (an oddly named SKU that combines both the Ti and SUPER brand extensions), considering that the current RTX 4070 Ti maxes out the AD104 silicon (60 SM), 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. It remains to be seen if NVIDIA utilizes the full 256-bit memory bus. As for the RTX 4070 SUPER, there exists a rather wide gap between the current RTX 4070 Ti and the current RTX 4070. While the RTX 4070 Ti uses all 60 SM on the AD104, the RTX 4070 only uses 46. This gives NVIDIA the opportunity to get closer to the RTX 4070 Ti in SM counts, and bolster against AMD's Radeon RX 7800 XT.
Sources:
hongxing2020 (Twitter), VideoCardz
There is no word on when the company plans to release these, or what their specifications are, but we can certainly speculate. The current RTX 4080, while based on the AD103 silicon, doesn't max it out—it uses 76 out of 80 SM (streaming multiprocessors) available on the silicon, but we doubt if those extra 4 SM could drive up enough performance to make a whole new SKU, especially given that the 256-bit memory bus of the AD103 is maxed out. We predict that the RTX 4080 SUPER could be based on the larger AD102 silicon that physically has 144 SM that the current RTX 4090 uses 128 out of. NVIDIA has the opportunity to pick an SM count such as, say, 96. AD102 also has a wider 384-bit memory bus, giving NVIDIA the option of either giving the RTX 4080 SUPER the same 24 GB memory configuration as the RTX 4090, or even 20 GB, across a 320-bit memory bus.
76 Comments on NVIDIA Readies GeForce RTX 4070 SUPER, RTX 4070 Ti SUPER, and RTX 4080 SUPER
The point is the only generation with "Super" SKUs had no "Super" SKU faster than their non-Super predecessor by more than 15 percent. We're currently into 2023 and even 5 percent FPS/$ gain is enough to make "Super" Ada SKUs more feasible than their RDNA2/3 competition all around. 8 percent would be the ultimate hard counter. There is no point for Huang to do anything better than that.
Ada's been around for a year, if a refresh isn't released within the next 3 to 6 months, it may be a generation that won't receive one altogether. I fully expect Blackwell 50 series to ship within 1 year to 18 months from now.
Also not sure if anyone brought this up, but did JHH have this in mind when of thought of RTX ~ Resiniferatoxin
Of course, GPUs are measured and quantified by a bunch of numerical metrics (cores, frequencies, etc.) so Nvidia really can't just slap the SUPER moniker on the exact same part. About the only given is that the SUPER SKU will slightly outperform the normal model and that it will be more expensive.
That's pretty much all anyone sane can expect. LOL, maybe the only honest thing the SUPER adjective is good for is the price.
:):p:D
Is this a joke?
They have dies that didn't qualify as a full 4090, so they're going to bin them as a 4080Super.
They have dies that didn't qualify as a full 4080, but are better than a 4070ti, so they're making a 4070Super.
They have dies that didn't qualify as a full 4070, but they're better than a 4060ti, so they're making a 4060Super.
There is even more fine-grain examples than just these as well.
This is a binning practice we've all seen for decades. It's not new and if we understand what is really going on, not at all surprising. It's just standard business. Nature of the machine, as it were.
the binned 102 are in RTX 5000 Ada and those are not necessarily a bad batch of dies just laser cut.
Perhaps this is for the mobile line up as there is a huge performance gap (like 40% to 50% faster) between the RTX 4070 and RTX 4080. Maybe this is where a 4070 Super would fit in.
Trying to insert a Ti SUPER is likely just to do more brand damage going into the future.
The supers will bring the interest back.
It doesn’t matter if they are good or vfm.
All in all it all comes down to price.
www.techpowerup.com/gpu-specs/rtx-5000-ada-generation.c4152
The Ti Super is a new creation though.
This is the problem with Nvidia selling their lowest-tier part with a mid-tier name, they've f*cked the naming scheme for refreshes. Unless they use AD106, there's no way to stop the vanilla 4060 Super from being a waste of effort, as it'll just be a 5-10% clock speed bump at the cost of efficiency.
Back in the day you had the vanilla, ti and ultra.
Also liked the vanilla, gt and gtx.
4080 super ti ultra extreme might be new and great for marketing?
There is utterly no point in trying to make sense of consumer PC hardware naming especially when it comes to gaming. The companies know they are marketing to idiots, especially in the DYI crowd, and so it makes no sense at all. It's all just marketing gibberish for e-peen. Because most people who buy a GPU just go out and buy the most expensive nvidia card they can afford with the most convoluted name.