Lossless cutting of video requires the first frame to be the I-frame, but there is no simular requirement for audio. How does that work?
I want to later do looping of a video, so I'd like to know how to properly encode it for that, so that I can jump to a specific point without any delays. With video it's oblivious: you can jump to any I-frame, you shouldn't jump to any other frames. But how does that work for audio? In what chunks does the audio gets stored?
Also, how to make sure there are no audio clicks when you jump from somewhere to somewhere?
I want to do a simplistic video player later, that can only play a limited subset of formats. Something I will be able to understand.
AFAIK, audio slices are very short, so you won't loose anything audible. No clicks either. Do some tests !
Quote from: viefi on December 09, 2017, 09:51:08 PM
how to make sure there are no audio clicks when you jump from somewhere to somewhere?
Clicks can be caused by an abrupt change in audio level at an edit point. You need to do a short fade-out/in to prevent this. ADM is a simple video editor. I don't know how it handles audio transitions in Copy mode.
Here's a cut at 00:09:59.999 in a main camera's video & audio tracks in Sony Vegas. Notice how the program has automatically inserted a 10ms audio fade-out and a 10ms fade-in at the edit point. The length of this auto-fade is set in preferences and could be as short as 1ms, if required.
(https://dl.dropboxusercontent.com/s/vdx2fh2wgits0ow/Vegas%20edit%20point.jpg)
Longer fades can be manually performed.
Dan.