Nvidia’s Deep Learning Super Sampling Just press version 4.0 and it is the most impressive version so far. The flagship function is the generation of multiple frames, which made a great splash with the Debuted from RTX 5090 – Together with Great performance claims – But it is not the only one. DLSS 4 also presents other features of neuronal representation, better compensation and an important review of how all its RTX characteristics are handled, using a new “transformer” in real time, which uses the same type of architecture as the modern large language model , as chatgpt.
All this makes DLSS 4 more powerful than ever and is available on all RTX graphics cards, although with the generation of blocked pictures for the latest generations of NVIDIA cards, At least for now.
Multiple DLSS generation
The historical characteristic of the last DLSS is the generation of multiple nvidia paintings, a technology that works with AI that can generate up to three additional paintings using AI, based on a single rendering frame of the GPU. This is what allowed Nvidia to launch its RTX 5090 with such monstrous performance of “double RTX 4090”, and reach more than 200 paintings per second (FPS) in games like Cyberpunk 2077 and Black Wukong mytheven 4K with enabled ray tracking.
There are some warnings for the generation of multiple frames, namely, which needs at least 60 fps (even if the improvement of DLSS) is used to avoid some of the worst artifacts that can come from the AI that tries to generate frames without sufficient information. There may also be a strange ghost, or an effect similar to the light of the movement sometimes, so the 4x frame generation configuration is one that will probably see less use than 3x or even 2x, which can provide considerable elevation in the frames per second. .
The generation of paintings and the generation of multiple frames in particular, also introduce an additional admission latency, but the new reflex technology 2 of NVIDIA does a good job when balancing that.
Unfortunately, for now, DLSS Multi Frame Generation is an exclusive feature of the RTX 50 series, such as RTX 5090 and 5080. It is said that these are older graphics cards such as the RTX 40 series and even RTX 30-series 30 with Optimizations, but that is very up in the air.
DLSS 4 Super Resolution
The central characteristic of DLSS is its elevation of super resolution that allows the game to be presented in a lower resolution to save GPU power, and then it has AI’s frames to look like a higher resolution. It comes in a variety of quality modes depending on the user preferences, and there is also the option of a deep learning anti-aliasing that only compensate for additional quality, without improving performance. It also works with each NVIDIA RTX graphics card, from the RTX 20 series to the last RTX 50 series.
Regardless of the NVIDIA RTX GPU you have, you can take advantage of the DLSS 4 improvements for the Super DLSS resolution. The new transformer model improves the quality of the image DLSS dramatically on the previous design of neural conviction networks (CNN), allowing an increase in the number of parameters sampled from the rendering frame, which gives DLS more information for work in the image. That in turn reduces ghost and visual artifacts, improves movement resolution and softens the edges much better than before.
This transformative model also improves the effect of additional RTX technologies such as ray reconstruction, making the ray layout more effective in finer objects with visual artifacts of the viewer and less noise.
Better yet, this is only the first generation of the NVIDIA transformer model. He spent six years refining his CNN model, so we should see continuous improvements of DLSS image quality in the coming years, that all NVIDIA RTX card owners, regardless of the generation.
What games do DLSS 4 use?
One of the greatest defects of previous DLSS generations is its lack of early support. When the first version of DLSS was launched with the RTX 20 series cards in 2018, it was so new and fresh that only a handful of games did to us. In later years, we have seen hundreds of game developers adopt DLSS technology, along with competitive AMD and Intel standards: FSR and Xess.
However, with DLSS 4, Nvidia began to run. On day 1 of its RTX 5090 available, it had more than 75 games that supported DLSS 4 ready to play. You can find The complete list hereBut the highlights include Indiana Jones and the Great Circle, Stalker 2, Alan Wake 2, Black Myth Wukong, Cyberpunk 2077, And more.
In games without native support, Nvidia now offers DLSS cancellations through the NVIDIA application. This will allow you to force the Transformer model, as well as the strength quality configuration such as Ultra Performance and DLAA that are not natively included in each DLSS game.
Get more frames without dlss 4
If you are not configured to buy an RTX 50-series graphics card or do not have a NVIDIA GPU at all, you will not be able to enjoy DLSS 4, or any version of DLSS for the case. You can still use FSR and Xess in compatible games for some improved frame speeds, but what about the generation of paintings?
While the “RTX” version of Nvidia de ESO remains blocked in your own hardware, you can use Scale without losses To enjoy the generation of paintings in almost any game. It is a $ 7 application in Steam and we are great fans. Check it if you get a little fomo of all new RTX 50 news.
