So in this case, all things being equal, I save it as a CBR MP3 192 KBps. I've compared the differences in file size and the MP3 audio file size is exactly identical to the AC3 audio file size. So I feel fairly certain that I am not loosing any quality here.
I guess that makes sense in theory, but it seems strange that a 448 KBps 5.1 audio track would have a lower bit rate per channel than a 192 KBps stereo audio track. I don't know all the technical details about how all the various channels in a 5.1 audio track are combined to form a stereo audio track (and I don't care) to really know how to calculate what the stereo track's bit rate needs to be to accommodate for the conversion.
So that's the question: When converting a 448 or 384 KBps AC3 5.1 audio track to a CBR MP3 stereo audio track, what should the MP3's bit rate be set at to match as closely as possible the original bit rate of the AC3 file so as not to loose any quality or create any extra over sampling bloat?
Assuming you downmix 5.1 to 2.0, use whatever LAME setting that generally brings you transparency with audio. -V4 should do well.
Quote from: moozooh on 18 May, 2010, 05:26:12 PMAssuming you downmix 5.1 to 2.0, use whatever LAME setting that generally brings you transparency with audio. -V4 should do well.Thanks for the response, and the info. AutoGK does use LAME 3.97, but it does it all automatically. So whether it's using a -V4 switch or not is out of my control. I know it downmixes from 48KHz 5.1 to 48KHz joint stereo qval=2. But all that I can do is choose MP3 CBR or VBR, and a corresponding bit rate. I'm not sure what audio transparency means, or what it does. Is there a minimum bit rate size that allows this feature to be enabled that I should be using? Thanks again.