Subject:Question about DC Offset in a Stereo File

Posted by: Angels

Date:3/7/2013 11:20:06 AM

I was wondering how the DC Offset correction is calculated in a Stereo File. I was checking statistics and the "Average" value was fairly high for this particular stereo file; DC Offset corrected things but I want to be sure of "how".

Assuming two different DC offsets are found in each channel of a STEREO file, how is DC offset adjusted? Technically it should be adjusted to bring the AVERAGE of both channels as close to zero as possible in order not to affect imaging, as opposed to correcting each channel separately.

There's no mention of this in the manual and nothing popped in a forum search. TIA for any additional info.

Subject:RE: Question about DC Offset in a Stereo File

Reply by: musicvid10

Date:3/7/2013 11:28:47 AM

"AVERAGE of both channels" is the wrong term to use here.

I don't know "precisely" how SF does it, but I assume it is a Root Mean Square Regression, the kind we learned in Algebra II, and is quite accurate in this type of work. Yes, it is calculated independently, as many stereo recordings show an identifiable DC bias in only one channel. Understand that the bias does not affect the AC gain or phase, only the zero reference.

You can of course, apply the correction to one or both channels as you wish. Changing the bias in this way does not affect "imaging," as you call it; just makes it easier to apply effects and limiting / leveling uniformly in the editing space and reduce the possibility of inadvertent clipping. Think of the purpose as being to "center" the signal in the workspace.