I don't understand bitrates in AAC
Reply #9 – 2009-06-03 22:48:37
Oh, a lot to comment on!I think it all boils down to multichannel inefficiency of the AAC codec What inefficiency? I don't know any codec which is more efficient than (HE-)AAC for multichannel audio (or (HE-)AAC + MPEG Surround for very low bitrates). The problem might be the tandem coding as you describe it, i.e. coding with AC3 first, then re-encoding with AAC. That should be avoided anyway. So Ludo, in your case of 2.0 audio, AAC should work fine.In the Wiki, "-q" is VBR, with a number between 0 and 1, "-br" is ABR with a bitrate, and "-cbr" is CBR with a bitrate (really ? ^^). So it's a number between 0 and 1 OR a bitrate, but not the two of them. But in some audio profiles for MeGUI (a GUI to encode video and audio), and you said the same to me, the two appears for the same encoding... Did I? I thought I said the same as the Wiki you quoted. "-q" stands for quality. The nero encoder maps the quality value to a certain target bitrate range internally, depending on the profile (HE-AAC or AAC LC).Will I have artefacts if I use HE-AAC at 96 kbps Yes, but you will also have artifacts with AAC LC at that bitrate, for example, a reduced bandwidth (cutoff around 14 kHz), which will make the audio sound slightly duller than it originally was. Question is, which sound worse, the artifacts from HE-AAC or the ones from AAC LC? Just try both and see what you prefer.What's the AAC equivalent of LAME ABR 128 kbps ? HE-AAC 64 kbps ? LC-AAC 96 kbps ? Don't know for ABR, but for CBR, according to some old tests I cited here , it's LC-AAC around 100 kbps CBR. HE-AAC at 64 kbps in my opinion sounds worse than MP3 at 128 kbps.At the risk of having Schuyler grumble at me again, has anyone EVER used the high-complexity profile? Oh, I mean "main". Haven't heard of anyone. Is Schuyler reading this? Chris