Final Fantasy XV Benchmark Demonstrates The Benefits Of GeForce RTX and DLSS. Download Now

Our new, powerful GeForce RTX graphics cards boast an innovative new technology called Deep Learning Super-Sampling (DLSS), which combined deliver up to 2X the performance of previous generations. It’s coming to many games, 25 of which we’ve announced already.

One of those is Final Fantasy XV: Windows Edition, and now developer Square Enix has released a special version of the game’s benchmark that demonstrates how DLSS looks and performs in their graphically-advanced, performance-intensive, open-world role-playing action game.

If you’ve got a GeForce RTX graphics card, you can download this new DLSS benchmark here, though you can also run it without the benefits of DLSS on any GPU.

Running the benchmark gets you a score, but if you want to capture framerates, use MSI Afterburner’s or Fraps’ benchmarking functionality, or something similar. Performance from generation-to-generation increases by up to 89% with DLSS enabled, and on our latest RTX GPUs, gains of up to 40% over TAA are delivered by enabling DLSS, giving gamers even faster experiences, and extra headroom to crank up the quality of visual effects.

Final Fantasy XV: Windows Edition NVIDIA DLSS Benchmark Performance

In terms of image quality in the Final Fantasy XV DLSS benchmark, surfaces and game details are clearer, aliasing is reduced, dithering is almost entirely eliminated, ghosting of moving detail -- which can be seen with many popular AA methods -- is virtually nonexistent, and incorrect rendering of moving detail seen through transparencies is fixed. In other words, a clearer, cleaner, sharper image is observed.

Using interactive comparisons, we can take a closer look at the Final Fantasy XV: Windows Edition DLSS benchmark to demonstrate and exemplify several of these benefits.

First up, a look at the clarity of the scene, which is noticeably improved with DLSS, making foliage and surfaces clearer and sharper (hold ctrl and use the mouse wheel if you wish to zoom in or out):

Similar improvements to clarity can be seen throughout the benchmark

When the game’s playable characters get in their car, the improvements to anti-aliasing with DLSS can be seen on its shiny edges. And as the car moves, there’s zero blurring on the moving foliage seen through the windscreen when using DLSS.

TAA, in comparison, tends to blindly follow the motion vectors of the moving object, blurring the detail on the screen. DLSS is able to recognise that changes in the scene are more complex and combines the inputs in a more intelligent way that avoids blurring.

DLSS also generates superior results on the characters’ hair, which is comprised of alpha textures that suffer from dithering and temporal ghosting with TAA. With DLSS, dithering is barely visible, and there’s no noticeable ghosting.

DLSS also generates superior results on the characters’ hair comparison

Similarly, dithering is effectively removed by DLSS on the game’s dynamically-generated grass, which covers the terrain for much of the game’s run time.

dithering is effectively removed by DLSS on the game’s dynamically-generated grass comparison

If you want to try the DLSS benchmark for yourself, download here.

Deep Learning Super-Sampling: How It Works

Now you’ve seen how DLSS can improve image quality and performance, perhaps you’re wondering how it works?

First, we show a neural network thousands of screen captures from each DLSS game that are up to 64x Super Sampled (64xSS), and then another set of images captured without anti-aliasing. We then task the network with reviewing and comparing the shots, to learn how to approximate the quality of 64xSS without its immense performance cost.

As the network repeats the process its algorithms are tweaked, and eventually it learns to automate the process, delivering anti-aliasing approaching the quality of 64xSS, whilst avoiding the issues associated with TAA, such as screen-wide blurring, motion-based blur, ghosting, and artifacting on transparencies.

Of course, not everyone owns a supercomputer, so we package the trained data from our neural network into a small file that’s included with a Game Ready driver. With that, your GeForce RTX GPU will automatically know how to best render each DLSS game, ensuring optimum results each time.

Finally, we leverage Deep Learning Super-Sampling’s vastly-superior 64xSS- esque quality, and our high-quality filters, to reduce the game’s internal rendering resolution. This greatly accelerates performance, without a noticeably negative impact on image quality, as you can see in the Final Fantasy XV: Windows Edition DLSS benchmark.

This performance boost, combined with the improved rasterization performance of GeForce RTX graphics cards, sees framerates increase by up to 2X at 4K, compared to 10-Series Pascal-architecture graphics cards. And with this extra performance you can crank up the settings, whilst still enjoying a super smooth experience at 60 FPS or more, giving you the definitive gaming experience.

For more on DLSS, check out our game support article to learn which games are adding DLSS, our Graphics Reinvented tech deep dive, and the Turing Architecture Whitepaper. And for news of new DLSS integrations, stay tuned to GeForce.com. As for the full version of Final Fantasy XV: Windows Edition, we are working with Square Enix to integrate DLSS.