In this post slhck mentions that for VoD (not to be confused with live streaming) one should use one of the two methods for compression:
- Two-pass CRF
- ABR with VBV-constrained bitrate
I feel pretty confident in understanding how these methods differ, but I do not really understand if one is better than the other and if so why.
In my head, it makes the most sense to use a two-pass CRF with -maxrate (constrained mode). The encoder will then calculate the best bitrate for me, and if that bitrate is higher than, say my upload speed, the encoder will limit the bitrate. Because of this it makes no sense to use a fixed bitrate like setting -b:v 5000K unless it is super important to maintain a certain bitrate at all times. But why would it?
Am I wrong in my assumption here?
Should I go for a two-pass CRF with -maxrate or is there a downside I am not seeing?
