The presentation of AMD FSR 2.0 technology raised many expectations and the truth is that there were many reasons for this. The giant Sunnyvale has confirmed this new version of its well-known scaling algorithm overcame great limitations. which was present in FSR 1.0, its inability to use temporary items.

With FSR 1.0, scaling was generated taking into account only spatial elementssomething that simplified the workload that this technology entailed and allowed for very significant performance improvements even in less powerful graphics solutions, but this ultimately had a very large impact on image quality.

I have had the opportunity to try this scaling technique in several games and with different graphics cards and I can confirm that When you switch from ultra quality mode, you will start to notice such a large loss of quality which just doesn’t pay off.

The power mode represents such a huge loss of sharpness and image quality that it doesn’t make sense. I started this topic in a technical analysis of Dying Light 2, where I attached videos showing the impact of FSR and DLSS in image quality.

AMD FSR 2.0: Higher image quality at the cost of less performance improvements

FSR 2.0 technology works, as we expected, with temporal and spatial elements, which means that it takes into account information from previous images and offers a more accurate and better scaling. This translates into a much higher result than we could get with FSR 1.0 technology, it does not support the discussion, but it is clear that It was not a “free” improvement and, as a result, it represents an additional workload thanks to which FSR 2.0 offers less performance improvement than what we get with FSR 1.0, regardless of the graphics card we use.

That’s why there’s a loss of performance compared to FSR 1.0 and this also occurs on next-generation graphics cardssuch as the powerful Radeon RX 6800 XT and GeForce RTX 3080. I don’t want to embark on technical complexities because I want all our readers to understand this without any problems. Keep in mind that when using a scaling algorithm such as FSR 2.0, all of this happens:

• Reduced GPU load due to cropped rendering resolution. Depending on the mode used, the total number of pixels can range from 67% of the original resolution to only 33%.
• Although the workload is reduced in this respect, a new task derived from the application of the algorithm is introducedand this consumes rendering time on each frame. If you only need to work with spatial elements, the process will be completed before you have to add time elements as well.
• Due to does not use specialized hardware to accelerate the load represented by the algorithm, the impact on performance is ultimately greater than, for example, the second generation of DLSS, although it is true that the difference is small, at least for Deathloop.

Realistically, we now have only one game to compare, Deathloop, so I’d rather wait until I can make a bigger comparison so I can draw my own conclusions and with my own tests. However, it is clear that by adding this layer of work with temporary elements, AMD has clearly improved the quality of FSit is also clear that she did so at the expense of performance.R2.0, but

AMD did an excellent job and cut, but Second-generation NVIDIA DLSS technology remains a big opponent to beat, both for image quality and performance, at least until we see what Intel XeSS is capable of. In the attached video you will find a complete comparison of HardwareUnboxed.