You’ve probably heard the term NVENC from NVIDIA on a number of occasions, especially when it comes to video encoding , whether it’s streaming or rendering. In this article we are going to tell you what NVENC is , what it is for and what benefits it brings to those who use it.
As its name suggests, NVENC is a proprietary technology from NVIDIA as its name comes from the acronym NVIDIA Video ENCoding, no more, no less. Therefore, one of the requirements to be able to use it is to have a graphics card of the brand, but as we are going to explain below, not just anyone will work.
What is NVENC?
It is a feature in the brand’s graphics cards that is used to encode video, unloading this task that supposes a rather intensive computational load to the processor and making it the GPU , with many more cores, in charge of doing it. It was first introduced with Kepler architecture GeForce GTX 600 Series graphics cards, and therefore one of the requirements to use it is to have one of these GPUs or more modern.
The NVENC encoder is natively compatible with many streaming and recording programs such as Wirecast, Open Broadcaster Software (OBS), and Bandicam, and also works with Share game capture that is included in the GeForce Experience software.
GeForce graphics cards on the consumer market (non-professional) only support 2 encoding video streams at the same time, regardless of the number of graphics cards installed, so even in Quad-SLI configurations you can only encode two videos simultaneously. For its part, professional graphics support up to 21 simultaneous transmissions per graphics card, depending on the model and the compression quality chosen.
NVENC is accompanied by NVDEC (NVIDIA Video DECoder) to do the opposite operation, complementing each other. In the following diagram you can find all the information related to it.
The first generation of this technology came with the Kepler architecture, and since then it has been increasing at a rate of almost one version per generation. Thus, the second generation arrived with the Maxwell GM107 GPUs, the third with Maxwell GM20X, the fourth with Pascal GP10X, the fifth with Volta GV10X and Turing TU117, and the sixth generation with Turing TU10X / TU116. The seventh generation of NVENC is expected to arrive with Ampere, incorporating further enhancements and more performance.
What benefits does it bring?
As we explained just now, NVENC allows you to do hardware encoding and decoding directly on the GPU, freeing the CPU from this computationally intensive task. By having many more cores at your disposal and which are also optimized to handle video information, you not only get dramatically higher performance but also more efficient in terms of performance per watt.
The good part of this system is that most video encoding software supports it natively, so as users we will only have to select it in the options of that program. For example, in Open Broadcaster Software we can choose to process the video through NVENC in the options.
For users who have an NVIDIA graphics card, being able to use this feature will free up many resources to the processor allowing a better performance in it, but it must be taken into account that this task passes to the GPU, and for that reason the graphic performance can be seen penalized if the graphics card is not powerful enough.