<< Click to Display Table of Contents >> Navigation: Configuring the Surveillance Client > Video Decoding via GPU > NVidia |
For more information about NVidia graphics card compatibility visit this link: https://developer.nvidia.com/video-encode-decode-gpu-support-matrix#Decoder
NVidia GPU decoding is only available in the 64-bit Surveillance Client (Surveillance64.exe) and requires the installation of official NVidia drivers: https://www.nvidia.com/Download/index.aspx?lang=en-us
It is possible to use 2 or more video cards for decoding, and the system will balance the decoding load between the available video cards.
The system also implements a video card resource monitor, where it will constantly evaluate GPU usage and GPU memory usage. If the GPU or memory usage is above the configured limit, the system will offload and decoding via CPU (Software) will be used.
Each camera instance being decoded (Regardless of the resolution used) will use around 200~250MB of GPU memory.
•H.264 Decoder: Enables H.264 video decoding via NVidia GPU.
•H.265 Decoder: Enables H.265 video decoding via NVidia GPU.
•Parallel Decoding: The system allows parallelism for H.264 and H.265 video decoding via GPU. This option can be used to speed up video decoding on the client, especially ultra megapixel images. Using this option will add at least 1 frame of delay to the video, that is, at 30 frames per second the additional delay will be at least 33ms while at 7 frames per second the additional delay will be at least 143ms.
•Maximum Memory Usage: Select the maximum GPU memory usage value for decoding. If maximum memory usage is reached, new cameras will be decoded via software.
oDefault: Restores the default value for this option.
•Use GPU Usage Manager: Enable this option for the system to monitor GPU usage and start offloading cameras for software decoding if GPU usage is above the configured limit.
oDefault: Restores the default value for this option.