Video Production Stack Exchange is a question and answer site for engineers, producers, editors, and enthusiasts spanning the fields of video, and media creation. It only takes a minute to sign up.
Sign up to join this community
I got into live streaming recently and while I was reading up on encoding settings I noticed that even though there are rules of thumb, there doesn't seem to be a standard set of bit rates we could point to for various requirements of quality like audio (eg 96 128 192 224 etc).
Why is that? Wouldn't a standard set of numbers be handy? Am I looking at this the wrong way?
Thanks
Complexity is the main thing. Because of how audio works, there tends to not be a whole lot of variation in the quality you get from a given bit rate. While video often has frames that are very similar from one image to another, audio rarely has the exact same sound playing for a long period of time.
This lack of predictability in audio means that the compression ends up being similarly efficient between one clip and another, so the quality levels can be closely associated to a given bit rate.
Video on the other hand can be predictable or unpredictable. Some videos sit on the same background for a long time with slow changes between scenes and large areas of similar color within those scenes. Such videos can be compressed very heavily and maintain their quality since large areas of the same color that remain consistent over time compress very well.
Other videos may have lots of action on complex background with rapid cuts between shots. Such video requires a much higher bitrate for the same quality because the rate of change within the video is much higher.
This distinction makes video unsuitable for "standard" bitrates because they are no guarantee of quality. Instead, standard metrics are based around arbitrary, subjective quality levels of video. Since there isn't a good concrete way to measure quality level, each company ends up using their own definition of what quality their videos are and use the balance of file size and quality to differentiate their offering.
The reason audio has 'standard' bitrates is because it evolved in such a way that was supported by the 'infrastructure' of the industry - that is, digital audio became a standard because those creating the content (record labels) decided "this is how CDs will work".
Digital video, however, has always* been a free-for-all, with each camera manufacturer creating their own proprietary format. As such, there has been no 'infrastructure' standard for video - the closest we have yet have been codec standards, not bitrate.
*Close enough
The bitrate for music has to do with the way our brain perceives audio.
I won't go into detail, but for each bit you add to a sample, you lower the noise floor -6dB. This means, that there is definable sweet spot for when humans starts to enjoy listening because the fidelity of what we hear through the speakers is close enough to what we think we would hear if the original audio source was there with us.
I guess I'm not the only who has confused a lamp for a person in the dark, or been fooled by visual illusions.
Our brain is funny this way, and much more sensitive to audible distortion, than to visual.
