Hi Bill (Grannyring),
I’ve had occasion to examine the waveforms and associated statistics of tracks from about a dozen or so CDs, encompassing pop, rock, and classical music, using a professional audio editing program (Sound Forge). Most did not have a single sample that reached full scale (i.e., the maximum possible digital value). A couple of them did, both being classical works having very wide dynamic range, but even in those cases that only occurred for a handful of brief instants during the course of a lengthy work. While on the rest of the CDs the maximum instantaneous volume of any sample ranged between a fraction of a db below full scale and many db below full scale. And I would expect in cases where the maximum sample value comes close to full scale that the original captures were made with much greater margins, and levels were subsequently boosted during the mastering process with software assuring that the maximum sample value would not reach 0 db.
In general, I would expect that with the possible exception of some works having extremely wide dynamic range any reasonably competent recording engineer would make a point of not over-driving (exceeding the maximum possible digital value of) the A/D converter.
So I would not expect the issue referred to in the ICC writeup to be much if any concern on the vast majority of recordings.
Also, the major contributor to what is referred to as the "loudness wars" is not what is described in the writeup. That term is usually used to refer to dynamic compression that is intentionally applied in the engineering of the recording, to make soft passages louder and loud passages softer. That creates the perception of a louder recording than if dynamics were not compressed, and makes the recording better suited to noisy environments, such as cars, and more attention grabbing in such environments.
So the bottom line seems to me to be that **IF** the ICC is implemented such that it does nothing unless samples reach 0 db (full scale) it will make no difference on most recordings, and only an occasional difference on the others.
Best regards,
-- Al
I’ve had occasion to examine the waveforms and associated statistics of tracks from about a dozen or so CDs, encompassing pop, rock, and classical music, using a professional audio editing program (Sound Forge). Most did not have a single sample that reached full scale (i.e., the maximum possible digital value). A couple of them did, both being classical works having very wide dynamic range, but even in those cases that only occurred for a handful of brief instants during the course of a lengthy work. While on the rest of the CDs the maximum instantaneous volume of any sample ranged between a fraction of a db below full scale and many db below full scale. And I would expect in cases where the maximum sample value comes close to full scale that the original captures were made with much greater margins, and levels were subsequently boosted during the mastering process with software assuring that the maximum sample value would not reach 0 db.
In general, I would expect that with the possible exception of some works having extremely wide dynamic range any reasonably competent recording engineer would make a point of not over-driving (exceeding the maximum possible digital value of) the A/D converter.
So I would not expect the issue referred to in the ICC writeup to be much if any concern on the vast majority of recordings.
Also, the major contributor to what is referred to as the "loudness wars" is not what is described in the writeup. That term is usually used to refer to dynamic compression that is intentionally applied in the engineering of the recording, to make soft passages louder and loud passages softer. That creates the perception of a louder recording than if dynamics were not compressed, and makes the recording better suited to noisy environments, such as cars, and more attention grabbing in such environments.
So the bottom line seems to me to be that **IF** the ICC is implemented such that it does nothing unless samples reach 0 db (full scale) it will make no difference on most recordings, and only an occasional difference on the others.
Best regards,
-- Al