AMD released Radeon Super Resolution or RSR for short in March 2022. RSR came with Radeon’s Adrenalin Editon driver update. RSR promises to speed up your games with driver-level technology. allowing you to change resolution below native fullscreen.
Now you are probably wondering don’t AMD already have an upsampling technology known as FidelityFX Super Resolution (FSR)? Well, you are right AMD does, but there is some key difference between the two technologies that we will go over later on in the arctic.
AMD is not the only company that is releasing upsampling technology. Both Intel and Nvidia have their own technologies that promise the same benefits. These technologies’ sole purpose is to use software to boost the hardware performance of your GPU. Every PC gamer’s mission is to match high frame rates with graphic output. These technologies are there to help you with that.
Nvidia has DLSS ( Deep Learning Super Sampling) Which helps boost your frames ingame. by rending the game internally at a lower resolution and using artificial intelligence to keep graphics visually sharp and upscaling the resolution. So there isn’t a noticeable drop in visual quality. Intel on the other hand are using XeSS technology which they will be releasing with their upcoming GPU lineup. It works in a similar way that DLSS does.
The main difference between DLSS and FSR is that Nvidia requires graphics cards to have RT and tensor code hardware. Currently, only the new RTX series has. FSR on the other hand doesn’t need this requirement and will work on a wider range of GPUs.
What is the difference between FSR and RSR
The major difference between FSR and RSR. Is that RSR implemented on a driver level. Meaning it can be supported on any game that has full-screen output. FSR on the other hand leaves the support to the developers of the games to integrate the technology with their selected titles.
Secondly, RSR has the ability to support more games since it is a driver-level update. It is limited by hardware. RSR will only work on Radeon RX 5000 and RX 6000 GPUS and greater. FSR has a wider range of hardware support since it is supported by the title that you are playing. FSR will even be supported on non-Radeon cards such as Nvidia Geforce RTX 2080 Ti.
As I previously mentioned before FSR leaves its support to the game developers. With that being said it is superior as far as graphic output. RSR does us the same underlying algorithm. But FSR is able to focus better on finer details that RSR can’t achieve because it is driver side. For example menu items and text will look better with FSR because the upsampling is normally done before screen overlays containing text are in place. RSR on the other hand will sample the text and overlays as it is working on a per-frame basis. HUD displays are a key overlay that will look better in FSR but won’t get the same quality of RSR.
Finally, RSR performance depends on the resolution that you are using. For obvious reasons, if you are using a 1080p and upscaling to 4k there should be a more dramatic increase than upgrading from 1440p. Now you will get a greater increase in performance if you are using 1080p upscaled to 4K resolution with an increase of around 2.8x if you are 1440p then you will get a 1.8x performance increase. As demonstrated in the picture below.
What should I choose FSR VS RSR?
If you are looking for a simple answer FSR is generally superior to RSR. Because there are more elements on the screen that are being more accurate upscaled while using FSR. But remember you can’t us both. Running both technologies will cause conflict. So if the game that you are playing supports FSR I would recommend using FSR over RSR.
Remember if you are using you’re using an older Radeon GPU than an RX 5000-series graphics card or greater. RSR technology is still great to us. If the game that you running doesn’t support FSR then go ahead and use the tech. Using upsampling is always going to be a better option to get the wanted FPS boost and graphic quality.