Skip to main content

Nvidia’s RTX Super might be a massive boost at no extra cost

The PNY RTX 4080 XLR8 installed in a PC.
Jacob Roach / Digital Trends

We just got another round of leaks surrounding the rumored Nvidia RTX 40 Super lineup. This time, the sources speak about the amount of memory these new GPUs might serve up. Nvidia may be making progress toward rivaling some of AMD’s best graphics cards, which typically have a lot more VRAM. But is this upgrade really going to be as significant as some speculate? Not all reputable sources agree.

A third source has just weighed in on the matter of the rumored RTX 4080 Super, RTX 4070 Ti Super, and the RTX 4070 Super. Benchlife cites its own sources at Nvidia’s board partners as it claims that the RTX 4080 Super might come with 20GB of memory, whereas the RTX 4070 Super might be boosted to 16GB. Doing so would definitely require chip changes for both cards. The RTX 4080 Super would now share the same AD102 GPU as the flagship RTX 4090, and the RTX 4070 Super would be bumped up to the AD103 chip found in the RTX 4080.

These are sizable upgrades that we’re talking about here. The RTX 4080 Super would receive 25% extra memory and a wider 320-bit bus. Switching to the AD102 chip also opens up the possibility of an increase in CUDA cores. Overall, this GPU would bridge the gap between the RTX 4080 and the RTX 4090.

The RTX 4070 Super marks an interesting change. If it does switch from the AD104 to the AD103 chip, it might end up with better specs than the RTX 4070 Ti, which runs on AD104 and has 12GB of VRAM across a 192-bit bus. Where does that leave the RTX 4070 Ti Super, if that’s going to be a thing? Most likely also only the AD103 GPU, which makes this a pretty crowded chip. We could end up with three graphics cards where the performance jump from one to the next isn’t that big, although the gap between the RTX 4070 Super and the RTX 4080 should still be significant.

I have some different opinions.
I think the relationship between RTX 4080 Super and RTX 4080 is similar to that between RTX 2080 Super and RTX 2080. https://t.co/LJqKZRISta

— kopite7kimi (@kopite7kimi) October 23, 2023

While Benchlife’s report is interesting, well-known leaker kopite7kimi disagrees with these rumors. According to kopite7kimi, the refresh isn’t going to be as big a deal as the previous leaks suggested. This speculation is leaning more toward a minor enhancement that unlocks the full potential of the GPU and boosts memory speed. In fact, kopite7kimi believes that the RTX 4080 Super will max out the AD103 chip instead of being upgraded to AD102, and that change can’t possibly result in any major performance increase. The RTX 4070 Super is said to be a slightly cut-down version of the RTX 4070 Ti.

Seeing as the rumors have been consistent that the new RTX 4080 Super might cost the same as the current RTX 4080 ($1,200), the minor upgrade seems more consistent with what Nvidia usually does. We’ll have to wait and see.

Editors' Recommendations

Monica J. White
Monica is a UK-based freelance writer and self-proclaimed geek. A firm believer in the "PC building is just like expensive…
Nvidia’s DLSS 3.5 update is what ray tracing always wanted to be
Reflections on the street in Cyberpunk 2077.

Nvidia's Deep Learning Super Sampling (DLSS) is getting a huge boost. The new DLSS 3.5 update adds a feature called Ray Reconstruction to the suite, and it promises to make ray tracing more realistic than ever before. I've tested it, and Nvidia was telling the truth.

Ray Reconstruction brings ray tracing to new heights of realism, and it's a fantastic addition to the DLSS suite. Even better, it works across all RTX graphics cards, unlike Nvidia's DLSS Frame Generation. However, there could be a problem with support as we see more games release with the feature. Ray Reconstruction may work with any RTX GPU, but it could be a feature that's only realistic for intensive ray tracing that requires one of the latest and greatest GPUs.
What ray tracing should be

Read more
Someone tweaked AMD’s RX 7800 XT, and the results may surprise you
YouTube channel Techtesters reviewing an underclocked AMD RX 7800 XT graphics card.

AMD’s RX 7800 XT graphics card handily beats Nvidia’s RTX 4070, outperforming it in games and costing less to boot, with the GPU comparison falling firmly in AMD’s favor. The main drawback? Its power consumption is much higher. But it doesn’t have to be that way, as one YouTube channel has shown that undervolting the card could have a notable impact.

When set up right out of the box, AMD’s card has a power draw of 263 watts, while Nvidia’s offering sips just 200 watts. If you’re concerned about your graphics card getting hot and heavy in operation -- and spiking your electricity bill -- the RTX 4070 is probably the better choice.

Read more
Asus’ new RTX 4090 shattered GPU overclocking records, and you’ll be able to buy it soon
Asus ROG Matrix RTX 4090 graphics card.

If you thought that the RTX 4090 was already the best we'll see in this generation, Asus is here to prove you wrong. Nearly a year after the initial launch of Nvidia's top graphics card, Asus is about to release the ROG Matrix GeForce RTX 4090 -- a one-of-a-kind GPU that already has a new world record under its belt.

ROG Matrix GeForce RTX 4090 Graphics Card

Read more