Quote:
Originally Posted by karbomusic
Yes, integrated is the one that considers the entirety of the audio file.
For the contest we don't need it but academically... "Loudness RAnge", it is similar to Dynamic Range but it doesn't consider the bottom 10% or top 5% of the signal when it does that calculation so don't expect it to always match something like TT-DR. So thinking about that, if you extended your render by 30 seconds but that was silence (less than 10%), it wouldn't change the LRA, but if that 30 seconds were some instrument at say 11% volume, then the LRA would change because it would then take that long low-level signal into the LRA calculation.
The first one should be considering intersample peaks (aka reconstructed waveform). The second term is the same thing but tells how many of any ISPs went above 0 dBFS. So you can use "Highest Recontstructed Peak Level" as your -0.5 target (There is a possible exception but I don't think it matters unless someone corrects us).
I don't know too much about that other than whatever difference that makes, it applies to every sample in the song (well as far as the totality of the calculation goes me thinks). Meaning, I'm not familiar enough with how the algorithms work. It's still sort of sinking in after reading up on Saturday and could be missing something.
|
Okay, thanks again Karbo.