DLSS is an acronym for Deep Learning Super Sampling and it is an Nvidia RTX feature that uses artificial intelligence to boost a game’s framerate performance higher, coming in handy when your GPU is struggling with intensive workloads.

When using DLSS, your GPU essentially generates an image at a lower resolution to lessen the strain on the hardware, and then it adds additional pixels to upscale the picture to the desired resolution, using AI to determine what the final image should look like.

And as many of us will know, bringing your GPU down to a lower resolution will result in a significant frame rate boost, which is what makes DLSS technology so appealing, as you’re getting both high frame rates and a high resolution.

Right now, DLSS is only available on Nvidia RTX graphics cards, including both the 20-Series and 30-Series. AMD does have its solution to this problem. FidelityFX Super Resolution provides a very similar service and is supported on AMD graphics cards.

DLSS is supported on the 30-Series line of GPUs as the RTX 3060, 3060 Ti, 3070, 3080 and 3090 come with the second-generation of Nvidia Tensor cores, which offers greater per-core performance, making it easier to run DLSS.

Nvidia is also expected to announce its latest generation of GPUs during its September GTC 2022 Keynote, the Nvidia RTX 4000 Series, codenamed Lovelace. If you’re interested in watching the event as it goes live, make sure to check out our article on how to watch the Nvidia GTC 2022 Keynote.

While nothing has been confirmed yet, the RTX 4000 Series is likely to include the RTX 4070, RTX 4080 and RTX 4090. We expect that the Nvidia RTX 4000 Series will provide DLSS capabilities, potentially to a higher extent than its predecessor, though we will be sure to update this article once we know more about the Lovelace series and have reviewed them.

Does DLSS reduce visual quality?

One of the biggest criticisms of the tech when it first launched was that many gamers could spot that the upscaled picture often looked a little blurry, and wasn’t always as detailed as the native image.

Since then, Nvidia has launched DLSS 2.0. Nvidia now claims that it offers image quality comparable to the native resolution.

What does DLSS actually do?

DLSS is achievable as Nvidia has gone through the process of teaching its AI algorithm to generate better-looking games and how to best match up with what’s already on screen.

After rendering the game at a lower resolution, DLSS uses previous knowledge from its AI to generate an image that still looks like it was running at a high resolution, with the overall aim of making games rendered at 1440p look like they’re running at 4K, or 1080p games in 1440p, and so on.

Nvidia has claimed that the technology for DLSS will continue to improve, though it is already a solid solution for anyone looking to see significant performance uplifts without the game looking or feeling too different.