NVIDIA will release new graphics cards during the month of January 2024… before focusing on the next generation.
Barring a very unlikely last minute change, in just a fortnight, NVIDIA will present its GeForce RTX 40 series, “SUPER” models designed to give a little boost to its mid/high range.
Little by little, all the characteristics of these cards were announced through very solid rumors. The last missing technical information has just been published.
Faster VRAM on the RTX 4080 SUPER
These are our colleagues from VideoCardz who announce the thing and this final clarification concerns the type of video memory used by NVIDIA. Quite logically, it will be GDDR6X.
The (small) surprise comes from the fact that the three cards considered by NVIDIA will not be in the same boat. Thus, we will have on one side the most powerful card – the RTX 4080 SUPER – with GDDR6X in 23 Gbps while the other two – RTX 4070 Ti SUPER and RTX 4070 SUPER – will have to make do with GDDR6X in 21 Gbps.
While the interface buses also differ significantly, this lag in video memory involves incommensurate bandwidths. This is how we speak of 736 GB/s, 672 GB/s and 504 GB/s respectively for the RTX 4080 SUPER, RTX 4070 Ti SUPER and RTX 4070 SUPER.
Slightly increasing frequencies
Our colleagues are also able to confirm most of the other characteristics of these three new GeForce models which display a TDP of 320 W, 285 W and 220 W, always in the same order.
It is also interesting to note that the operating frequencies are increased compared to non-SUPER models. We gain 90 MHz base and 45 MHz boost on the RTX 4080 SUPER. On the RTX 4070 Ti SUPER it’s 30 MHz base and nothing at all in boost. Finally, the RTX 4070 SUPER progresses by 60 MHz base and… still nothing in boost.
The table above is provided to us by VideoCardz and you will note that there remains, so to speak, only one unknown, but it is important since it concerns the price of these cards.
Source : VideoCardz
1