Does the new technology fit DLSS?
Hey, God! Introduced with the launch of Radeon RX 6X50 XT series Version 2.0 of FidelityFX Super Resolution Technology launched, Hey, God! FSR 2.0. This technique has a similar purpose NVIDIA Deep Learning Super Sampling (NVIDIA DLSS), and the most recent Intel XE Super Sampling (Intel XeSS): Make it possible to render games at lower resolutions, use image enhancement techniques while maintaining maximum detail, and thus deliver greater performance with less loss of visual quality.
More than an evolution, FSR 2.0, FSR 1.0 . completely different technology than
Hey, God! FidelityFX Super Sampling 2.0 is its own “built from scratch” technology Hey, God!, so different from version 1.0 that it almost deserves its name. Let’s know the new strategy Hey, God!How results in performance and graphic quality as well as controversy NVIDIA DLSS Do we no longer need Tensor Cores and, consequently, RTX to do a good upscaling?
actually an amd dllss
The first major change from AMD FSR 2.0 is that it is now actually very similar to Nvidia’s approach to DLSS, or DLSS 2.0. AMD FidelityFX Super Sampling 1.0 did the same thing as all of the techniques listed so far, but with a different technique. He will take an image, enlarge it, and use FidelityFX Contrast Adaptive Sharpening (FidelityFX CAS) to retain image detail and definition.
New approach closer to DLSS
AMD FSR 2.0 now uses more information with a temporary filter. In other words: FSR 2.0 takes into account information from the previous frame when deciding what each pixel will look like. It will also use other information from the game engine, the speed vectorOr motion vectors, which will inform the technology of where things are and where they are going, again helping to make the right choices for sizing the image.
If You’re Having Deja Vu, You’re Not Wrong, Because You’re Probably Missing DLSS version 2.0 . from our article on And how well it is working thanks to the super sampling technology on the GeForce RTX side.
FSR 2.0 takes motion vector information, object depth information, and color, mixes it with data from the previous frame, and decides to fill in the missing blanks to increase the image resolution.
The approach of FSR is very similar to that of DLSS in its methods. Essentially you are downsampling the image, actually rendered, and you are using more machine learning to fill in the gaps in high performance mode. Below is a progression of these modes, remembering that they refer to the final resolution, or output resolution, in this table:
– After the advertisement continues –
DLSS vs FSR 2.0 comparison in video
FSR 1.0 vs. FSR 2.0 and vs. DLSS Graphics
Time to look at development versus FSR 1.0, and how is the contention with DLSS in terms of graphics. We compared them with entry and intermediate cards from both Nvidia and AMD, with 4K resolution and Full HD. For the first test, let’s look at the performance in 4K view:
Below we’ll analyze the FSR’s performance in this view, the quality progress in Balanced and Quality modes. To illustrate the difference, we’re using a 5x zoom to get the details:
The result of quality mode is excellent, It’s clear how it manages to retain a lot of detail, and thanks to the sharpening effect, we get the feeling that the resolution is even higher than the image actually rendered at 1080p. When turning to display mode, the image begins to deteriorate, showing more artifacts especially on moving objects, something that becomes more apparent when watching the comparison video. Despite the loss of quality, playing FSR 2.0 in performance mode is still quite viable, especially when you’re not with the “nose sticking out of the screen” that is the 5x zoom effect.
Now let’s go back to the comparison, but let’s put FSR 2.0 to compare with 1.0 and Nvidia DLSS:
– After the advertisement continues –
Implementing the 4K display mode of the three technologies, i.e., all rendering in Full HD and moving up to 4K, the evolution of FSR 2.0 is remarkable. The image has a lot more definition, retains more detail and has less artifacts at the edges of moving elements., DLSS still has an advantage, with more defined edges, less artifacts on objects, but the difference is already negligible, especially without the zoom we used in the comparison above.
Let’s move on to the second view, now to examine the difference between these techniques in their quality modes, i.e. from 2560×1440 to 3840×2106.
In this scene we have two zoomed segments:
In this part of the comparison, still images don’t do justice to the elements that are more apparent in the video. Both native 4K with TAA and FSR 1.0 have serious flicker issues at the edges of objects. FSR 2.0 and DLSS reduce or completely correct most of them, providing a much more finished and better image than native 4K rendering. DLSS has the advantage and handles certain patterns better, such as grids stuck on the roof of a building, but in this comparison, FSR 2.0 handled character movement in the background better. Here I recommend rewatching this scene than in the video.
Now let’s see what the FSR 2.0 technology looks like at a lower resolution. In this case, we go with the Radeon RX 6600, which handles games at Ultra, RT On, and Full HD resolutions.
Again, flicker is an element that will only be visible in video comparisons, but I have already emphasized that it was clear that FSR 1.0, even in quality mode, suffers greatly from this problem. Is. FSR 2.0 in performance mode had this problem as well, in addition to losing definition similar to what happens in FSR 1.0 quality. In short: Aim for the ideal FSR 2.0 quality mode for 1080p. The graphical result is great, and although the performance gain is modest, it’s more interesting than natively running and applying TAA.
This proves to be an unavoidable effect of working at very low resolutions. Like DLSS, FSR 2.0 suffers when the final resolution is too low. In Quality mode, it renders from HD to deliver Full HD, and in Performance mode we have an average of 890×540 as base resolution, which is very little detail for upscaling reconstruction.
FSR 2.0 Performance
We compared the performance of FSR 2.0 with version 1.0 of the technology and DLSS, which is implemented in version 2.3 of Nvidia’s technology in Deathloop.
Entrance and Old Signs
Starting the comparison with older cards, we see that the popular Radeon RX 580 and GTX 1060 6GB make a lot of gains in performance, especially the Radeon. Going to the latest cards, such as the GTX 1650 GDDR6 and Radeon RX 6500 XT, we get two opposite results.
The GeForce GTX 1650 GDDR6 benefited greatly from FSR 2.0, with a gain of 20%, while the Radeon RX 6500 XT practically stalled at a much worse frame rate. In fact, this card ran all tests poorly, indicating that this model’s limitations—which we’ve already gotten on our feet—have negatively affected this test.
AMD FSR 1.0 still delivers more performance gains, but as we show in the comparison, it’s not on the same level as the other two technologies.
Starting with Full HD cards like the RTX 3050 and Radeon RX 6600, we’ve had decent results enabling FSR 2.0 quality, and interestingly, GeForces made the most. The RTX 3050 has increased its performance by about 25%, with a gain of 15%, which is still good, but far less relevant. DLSS performed slightly better than FSR 2.0, while FSR 1.0 and its simple implementation remained the best test result.
And finally, we took a look at 4K with the RTX 3060, a card that doesn’t have the performance to handle games at that resolution. All technologies have given a respectable boost, with FSR 2.0 already bringing in a 30% gain and putting in an average of 50fps. This suggests that using only FSR in Performance mode, which still has a decent level of quality, or changing the preset from Ultra to High, makes 4K gameplay viable on the RTX 3060. But it’s even more game to take advantage of Tensor Core and enable DLSS, which will bring gameplay in high quality to 60fps+.
AMD FSR 2.0 still has some drawbacks compared to DLSS 2.3, But it also won’t require or defeat Nvidia’s resource binding. just get close To show itself as a legitimate alternative for Radeon owners. And he did.
The AMD FSR 2.0 didn’t even need to beat the DLSS, just get close enough. and he got
Version 2.0 of the FSR is comparable to DLSS in quality, and if it weren’t for our zoom-in reviews, it would have been hard to tell the difference. With the gameplay going on, it’s unlikely that anyone will notice a difference, unless you stick pixels under a microscope and stare. This is good news not only for AMD card owners, but also for owners of GeForces without DLSS support.
But it’s not all success, and AMD FSR 2.0 requires a lot of developer intervention. Fortunately DLSS 2.3 paved the way for this type of temporal filter with advanced reconstruction, so adoption will probably be faster than DLSS in its early days. But it still remains to be seen in practice.
Owners of DLSS enabled cards should continue to take advantage of their GeForce RTX’s tensor cores. DLSS is faster than AMD FSR, provides more frames and produces slightly better graphics results. Next, AMD FSR is a great option, and should be used in 4K even in display mode, while Full HD is ideal for use in quality mode. If that doesn’t provide enough frames to play with, then with FSR 1.0 is the way to go, which is even more interesting than not using the feature, in many cases it may be better to stick with the native resolution in these cases.
Source: amd community, GPU Open