XPost: rec.audio.pro, comp.dsp   
   From: gah@ugcs.caltech.edu   
      
   In comp.dsp Mike Rivers  wrote:   
   (snip on dBFS and such)   
   (then I wrote)   
      
   >> I sometimes record live high-school orchestra concerts.   
   >> Because it is hard to know the level, I record 24 bit, then   
   >> find the peak and RMS of each track.  Then I figure out how many   
   >> bits to scale each track by so that peaks stay below FS, and   
   >> they should sound about right together.   
      
   > I suppose that's one way of mixing when you aren't listening   
   > and don't have controls handy. From a computer standpoint,   
   > it makes a certain degree of sense, but from a musical and   
   > artistic standpoint, it's absurd.   
      
   Well, the idea is that they should sound about the same level,   
   such that one shouldn't want to run up and change the volume   
   control for each track.  I don't think I could do that very   
   well just listening to them, trying to memorize the average   
   level over a 15 minute track.  If the peaks are about the   
   same, and RMS about the same then I figure that they will sound   
   about right.  If both peak and RMS are different by about the   
   same amount (I might have changed the record level), then I   
   adjust by about that amount.  Usually that works.   
   For a recent recording, the four RMS values were -36, -39, -41,   
   and -35, all dbFS, peaks were -10, -11, -14, -11 dbFS, all   
   with the same record level.   
      
   -- glen   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)