How a supercomputer in NVIDIA HQ makes your gaming laptop faster

NVIDIA DLSS: deep learning supersampling. You know, we know, but let’s not pretend that everyone who plays games is also an expert in learning the incredibly complex AI algorithms. So what is it?

Introduced in the NVIDIA RTX 20 Series cards, DLSS is not only a big step forward for performance, but also a side step. Because, in addition to all the counting issues that are dealt with to death whenever you ask your gaming laptop to play the next frame, say, Control, Call of Duty: Black Ops Cold War or Watch Dogs Legion, some of these plays are enhanced by AI models created by a supercomputer many miles away from your trusted platform.

Which means the results on the screen are quite surprising for a small form factor card that powers a gaming laptop. Performance compensation must be done by mobile devices in the name of thermal management and the large amount of PCB that a designer had to work with. The exploitation of AI power, accelerated by tensor cores, is, however, a very literal game changer.

Of course, it is still anchored by the calculations that happen at death. This comes through the kindness of the AI ​​core of deep tensor learning. When presented with a complex image, the tensor core actually has the option to call a friend. And lucky for you, that friend happens to be a massive supercomputer. A supercomputer that has been specially trained to analyze the image with an ultra-high fidelity of 16K, compare it with the low resolution and create a model to interpolate the differences.

So when the core of your laptop’s GPU makes this call, basically what happens is accessing the AI ​​model compiled by the supercomputer and then written to an NVIDIA driver ready to play. This model allows the GPU to produce clear, high-resolution images much faster than traditional rendering.

Purchase now

What this means for you, happily looking for the ridiculously detailed frames that come to you quickly, is the ability to run a game at a higher resolution than you could otherwise without experiencing a drop in performance. Playing a 4K image natively requires a lot, but using DLSS to provide an image close to the same fidelity as native 4K requires only 25-50% of computing power – the result is significant performance gains.

To put it in real terms: vast firefights, people fighting in uniforms clear enough for a museum exhibit in Call of Duty: Black Ops the Cold War. Nameless scientific anomalies that reflect the sharpness of a marble floor in the mysterious office of the Control Agency. Adding it to the man from a dystopian London who looks, frankly, much better than reality through Watch Dogs Legion.

The initial implementation of DLSS trained the NVIDIA supercomputer only in the in-game assets, but since then DLSS has been updated and released on RTX cards and makes the Tensor core even smarter. Images outside of games are used to inform the deep learning AI how to analyze and sharpen an image, for a clearer frame on the gaming laptop screen, narrowing even more the gap between the original playback resolution and the game’s running mode. native at a higher level res. The result is a higher frame rate and increased performance.

Game performance is just the beginning. Broadcast features, such as the green screen RTX and RTX AR with NVIDIA Broadcast, use AI to provide streamers with tools to improve the way they interact with their audience with intelligent image recognition, which can accurately change backgrounds, shapes faces and can apply 3D effects on them. It also uses AI patterns to clean up background noises picked up by microphones.

And, of course, you can get all of this and more in HP-powered GeForce RTX gaming laptops – just click the link to take a look.

Source