Nvidia expands its proposal to gamers to improve the performance and definition of games. After DLSS, here is the return of Nvidia Image Scaling.
Perhaps shocked by the media attention paid to AMD FidelityFX Super Resolution, Nvidia has decided to update a technology already integrated in its drivers: Nvidia Image Scaling.
Behind this name hides a technique of spatial scaling of the definition, as a more general alternative than DLSS. Did you not understand this sentence? We will explain everything to you.
What is spatial scaling?
By releasing Image Scaling, Nvidia wants above all to differentiate between its flagship technology, DLSS, and spatial scaling engines. The brand is not wrong, there are fundamental differences between the two.
Spatial scaling is quite simply what we have known for a long time: the possibility of improving the definition of an image by stretching it artificially, like a digital zoom. There are as many recipes as there are software publishers to try to erase the faults and in particular the escalation effect that occurs with this practice.
AMD Fidelity FX Super Resolution, or AMD FSR is one such recipe. Nvidia Image Scaling is another. Your television has yet another built-in.
What is the difference between Nvidia DLSS and Nvidia Image Scaling?
Nvidia now offers two methods to improve the definition of a game in real time: Nvidia Image Scaling on one side and Nvidia DLSS on the other.
As the name suggests, Nvidia Deep Learning Super Sampling uses artificial intelligence to create the high definition image. Unlike a simple upscaling, DLSS can therefore create information, which explains why some games benefit from a better display quality than with native rendering.
According to Nvidia, the difference also comes from the source of the work. DLSS uses three images to work, the current image and the previous two, to understand the scene and create the high definition version, while Nvidia Image Scaling, the spatial scaling, uses only one image.
As in photography: the more data we have distributed for software processing, the better the end result will be.
What are the advantages of Nvidia Image Scaling?
We can therefore consider that Nvidia DLSS is more interesting than Nvidia Image Scaling. This is not surprising, this is already what our test of AMD FSR showed.
So what is Nvidia Image Scaling for? Like AMD FSR, it is about providing a solution for all PCs that cannot benefit from DLSS. You need a graphics card equipped with Tensor Core, the GeForce RTX, to use DLSS. AMD FSR and the scaling offered by Nvidia are open to any graphics card.
Obviously, Nvidia puts forward its own solution as offering better processing than AMD FSR. You shouldn’t expect miracles: the process is basically the same as AMD FSR, with the same qualities and the same disadvantages.
Remember that DLSS is an AI upscaling, which has proven to be better than traditional upscaling like AMD FSR or Nvidia Image Scaling. The other side of the coin: it requires special hardware, where traditional solutions work on any PC.
What games are compatible?
The other advantage of Nvidia Image Scaling is that it is compatible with a maximum of games, because it is a function carried by Nvidia in its drivers directly.
Obviously, Nvidia does not hesitate to point out that more and more games integrate DLSS. In particular, flagship games such as Fortnite, Battlefield 2042, the infamous GTA remastered trilogy, Deathloop or even soon the game God Of War.
Still, this is only a selection of compatible games, which do not represent the entire market, as popular as it is. For other games, this is where Nvidia Image Scaling comes in.
How do I activate Nvidia scaling?
Nvidia announces today the availability of the SDK for developers who wish to integrate the option directly into the game. Otherwise, it will be necessary to go through the Nvidia GeForce 496.70 drivers or more.
The image scaling option present in GeForce Experience makes it very easy to set the option for all games. We define the applied force of the algorithm to smooth the whole, that is to say the difference between the definition actually calculated by the PC and the definition displayed.
To follow us, we invite you to download our Android and iOS application. You will be able to read our articles, files, and watch our latest YouTube videos.