6 crucial GeForce RTX 40-sequence details you may well have missed

At very long last, the GeForce RTX 40-sequence has been discovered, but in legitimate Nvidia style, CEO Jensen Huang did not delve deeply into uncooked speeds and feeds although introducing the RTX 4090 and 4080 all through his GTC keynote. There is only so a lot time during these points, soon after all, and mass audiences really do not like sifting through verbal descriptions of technical footnotes. Thankfully, Nvidia’s site features crucial technological specifics lacking from the presentation—details that confirm several deeply exciting tidbits about these new “Ada Lovelace” GPU-driven graphics playing cards.

In this article are six should-know GeForce RTX 40-collection information that you did not listen to all through Jensen’s expose. For the complete scoop on Nvidia’s new graphics cards, examine out our coverage of the GeForce RTX 4090 and 4080 announcement.

One title, two extremely distinct RTX 4080s

Nvidia

In a deeply anti-shopper move, Nvidia is rolling out a pair of GeForce RTX 4080 offerings, ostensibly differing in memory capacity. Nvidia simply just gave them “16GB RTX 4080” and “12GB RTX 4080” names, after all. But when Huang failed to mention it, these two “4080s” will give incredibly distinctive baseline general performance as nicely.

As you can see in the chart higher than, the 12GB GeForce RTX 4080 doesn’t just have fewer memory, it also has a a great deal narrower 192-bit bus, and a whopping 21 p.c fewer CUDA cores. That usually means it will be considerably slower than the 16GB model, much more typically than not. Contacting these two really dissimilar GPUs by the similar identify is positive to result in confusion among graphics card consumers.

The 12GB RTX 4080 will not have a Founders Version model

Nvidia

Speaking of, scrolling by the spec sheet for the RTX 4080 unveiled this appealing supplemental tidbit. Although the 16GB GeForce RTX 4080 features unique duration, width, and slot measurements for Nvidia’s Founders Version edition, the 12GB GeForce RTX 4080 basically says “varies by company.” It does not audio like the 12GB model will get Nvidia’s own custom Founders Version procedure. Update: Nvidia’s Brian Burke confirmed to VFAB that there will be no Founders Edition edition of the 12GB RTX 4080.

GeForce RTX 40-sequence supports AV1 encode

Nvidia

AV1 encode is the holy grail for media creators, featuring significantly superior visuals at a lot decreased bandwidth needs. Intel conquer AMD and Nvidia to the punch by which include killer AV1 encoding in its Arc graphics playing cards, but the footnotes of Nvidia’s RTX 4090 and 4080 spec sheets expose that Team Environmentally friendly will assistance AV1 encode this technology as effectively (AV1 decode was currently supported).

“GeForce RTX 4090 and GeForce RTX 4080 graphics cards characteristic two of our new eighth-era NVIDIA Encoders (NVENC), now with guidance for AV1 encoding, enabling a raft of new alternatives for livestreamers, video clip editors and video callers,” the RTX 40-collection announcement article crowed.

You read through that right—not a single, but two NVENC encoders with AV1 encode assistance in RTX 40-collection graphics cards. Streamers, take take note.

DLSS 3 is not coming to more mature GPUs

It’s noticeable that final-technology GPUs won’t include things like new-gen components encoders. But Nvidia’s significantly-hyped DLSS 3 software package feature is not coming to RTX 20- and 30-collection playing cards possibly, even though DLSS 2. and 2.1 have been backported to both of those. Nvidia’s GPU era comparison website page lists DLSS 3 on the RTX 40-collection only though more mature-gen RTX choices continue to be listed with DLSS 2..

Nvidia

That’s due to the fact the DLSS 3 application requires components that more mature GPUs merely don’t have. Nvidia spokesperson Brian Burke stated the circumstance to VFAB via e mail:

“DLSS 3 is composed of 3 technologies – DLSS Frame Era, DLSS Tremendous Resolution (aka DLSS 2), and NVIDIA Reflex.

DLSS Frame Technology employs RTX 40 Collection large-velocity Optical Circulation Accelerator to work out the movement move that is employed for the AI community, then executes the community on 4th Era Tensor Cores. Support for previous GPU architectures would have to have further innovation and optimization for the optical movement algorithm and AI product.

DLSS Tremendous Resolution and NVIDIA Reflex will of study course stay supported on prior generation hardware, so current GeForce gamers & creators will benefit from video games integrating DLSS 3. We carry on to exploration and prepare the AI for DLSS Super Resolution and will give product updates for all RTX consumers as we have been performing because DLSS’s original release. For the broad the greater part of video game integrations, both equally DLSS Tremendous Resolution and Body Generations will be uncovered as individual UI configurations and a gamer can pick out to run one particular or the other or the two.”

You’re gonna will need a more substantial boat

The GeForce RTX 4090 doesn’t suck down 600 watts-plus like some early rumors suggested—at minimum not in Nvidia’s Founders Edition iteration—but it does arrive with a 450W ranking, with a bare minimum 850W electric power source recommended. That matches the flagship RTX 3090 Ti from last technology. Like we reported then: You are gonna will need a bigger boat. And by boat, we signify power source. It is not all dear news even though. Even though the RTX 40-sequence supports the 12VHPWR-pin connectors observed in new-breed ATX 3. electric power materials, you are going to also be able to use three or 4 regular 8-pin ability connectors if you by now have a burly PSU.

NVLink is lifeless to GeForce

Nvidia’s SLI technological know-how for multi-GPU setups has been lifeless for a though now, but the “NVLink” tech that supplanted it remained in final-generation’s RTX 3090 GPUs. No much more. The spec internet pages for the GeForce RTX 4090 and 4080s explicitly say that NVLink is not supported this era. Pour a single out.

That’s it for these buried treasures. We’re certain to discover even additional about the GeForce RTX 40-series in the coming days and weeks, but when once more, for the total scoop on Nvidia’s new graphics playing cards, test out our protection of the GeForce RTX 4090 and 4080 announcement.

Editor’s take note: This report was updated to consist of remarks from Nvidia’s Brian Burke.