Having the content transformed via filters, you can now encode it. Let us know if you need information about other filters. Notice that it will increase RAM/CPU usage so use it only if you need to do something you cannot do on GPU. Then add "hwupload_cuda" filter to put it back into GPU processing pipe.After that you can use regular FFmpeg filters.Add "format" custom filter with "nv12" value to set proper frame format.
Add "hwdownload" to get the frame from GPU. To add them, add a custom filter, set its name as mentioned and leave the value field empty. However, you can take the frame out of GPU and process it separately using "hwdownload" and "hwupload_cuda" filters. Notice that regular Scale filter from the UI toolbox will not work with GPU-decoded frames as well as other regular ffmpeg filters just because the processing is done internally in GPU. There you need to specify "NVENC-ffmpeg" in Decoder field. Once you create a new scenario, drag and drop a blue decoder element onto the dashboard. Now create a new scenario to start a new pipeline setup. If you already have Transcoder installed, please run these commands to upgrade the package: This way you'll make sure the transcoder was set up properly. Create some simple scenario with CPU transcoding (e.g. Create a transcoder license and subscribe for it. Install the latest NVidia drivers on your server. The basic steps to make NVENC working are as follows: We'll support other upcoming LTS Ubuntu releases as well. Notice that the described functionality is available on Ubuntu 18.04, 20.04 and 22.04 only. If you're not familiar with Live Transcoder, take a look at Transcoder documentation reference. In our example we'll use a stream which application name is "input" and stream name is "source". We assume you've already set up Nimble Streamer, it's been set up to get an incoming live stream and you've tested basic streaming. We'll show you how to set up this NVENC-powered processing chain. 
This is done using specific FFmpeg libraries which we use in addition to our own code.
Nimble Live Transcoder now allows building transcoding pipelines which are performed completely with NVidia GPU hardware acceleration. That caused extra resources usage to transfer processed frames among CPU, GPU and RAM. NVidia NVENC always fully supported in Live Transcoder for decoding and encoding but all filtering operations were performed using CPU. Live Transcoder for Nimble Streamer provides wide feature set for transforming live content using both software libraries and hardware acceleration.
0 kommentar(er)
