In reality, it’s nothing more than an NVIDIA marketing name that has always been known as Anti-Aliasing Super Sampling, but with the difference that in the process of reducing the resolution of the image, a Gaussian filter with up to 13 different coefficients is used. Process which is performed by a Computer Shader controlled by the graphics card driver and which is executed when it is necessary to encode the back buffer to the front buffer for the display controller in charge of sending the information to send it at the correct resolution.
It is therefore a technique that allows us to improve the image quality of our games if we have a graphics card with enough power to move the graphics to a higher resolution and even frame rate, but our screen does not meet these requirements.
What is DLDSR?
The acronym DLDSR stands for Deep Learning Dynamic Super Resolution or Dynamic Super Resolution through deep learning and is based on using Tensor Cores to run the resolution reduction algorithm instead of using the ALU FP32
It is not the first time that Deep Learning has been applied to post-processing operations and therefore with the image already generated. It is at this stage that the use of systolic arrays has proven to be much more efficient in both speed and power consumption compared to the use of a SIMD unit. This is why the DLDSR not only included the adaptation of the already veteran DSR to Tensor Cores, but also a series of additional algorithms at the screen space level, i.e. post-processing. Thus, the DLDSR also makes it possible to automatically apply a series of additional algorithms in games.
It must be taken into account that the raw computing power of Tensor Cores is much higher than that of ALUs in FP32, CUDA cores, since the ratio in TFLOPS is much higher. Additionally, they can work with lower precision values, which makes them ideal for dealing with the different color components and the alpha channel of each pixel.
SSRTGI
The Screen Space Ray Traced Global Illumination is also known as the “Ray Tracing ReShade filter” and has the ability to improve lighting and shadows in older titles, causing an instant visual improvement.
SSAO
Screen Space Ambient Occlusion enhances the appearance of shadows near object intersections in 3D space to give the scene a sense of visual volume through the play of light and shadow. SSAO is essential for giving visual appeal in games that don’t rely on ray tracing as light travels through space.
Dynamic field of view
Also known as Dynamic Depth of Field, it is a technique widely used in photography, it consists of blurring distant objects to give more importance to near objects. It is therefore a widely used technique for creating highly cinematic scenes and is used in many games as part of their art component.
The driver is in the DLDSR manager
The graphics card driver is the software in charge of communication between applications and the graphics hardware you are using. However, there are a number of functions that running applications can benefit from that are handled neither by the operating system nor by the application, but by the controller itself.
The advantage is that game developers don’t have to worry about adapting these new features. In the case of DLDRS applying the algorithm in games does not require in-game programming and therefore support is already standard on all NVIDIA RTX graphics cards. Of course, the DLDSR to be able to be implemented requires that the GPU has integrated Tensor Cores and therefore we will not be able to make it work in a GTX.
Is DLDSR the same as DLSS?
Although both technologies work using deep learning, they are two completely opposite concepts. In the case of DLSS, we render an image at a lower resolution to generate its version with more pixels on the screen via AI. In the case of DLDSR, this involves natively rendering an image at a higher resolution by the GPU to generate another at a lower resolution.
Another big difference with DLSS is that although NVIDIA uses the acronym DL in DLDSR, it is not a learning algorithm, as there is no need to train the algorithm for each game’s visual style. Simply, it is about moving the execution of certain algorithms from one type of units in NVIDIA RTX GPUs to others. This therefore guarantees the compatibility of the DLDSR for all the games that exist on the market and, therefore, of the implementation at the driver level.
The use of post-processing algorithms through specialized units for AI is something that has been used for a long time, not in vain, many image filters in Photoshop-style programs have a version that allows them to be executed in Tensor Cores, which makes them run faster and with much higher precision.
Table of Contents