So, out of some kind of curiosity, I had a look at the values for %replaygain_track_peak% in my collection, and it looks like only lossy files have a peak over 1.0. A lot of lossless tracks with 1.0, but not one over. Why is that? Have all of them been loudened or is it some kind of calculation problem due to interpolation used in lossy codecs? And doesn't a peak over 1.0 indicate clipping?
I'm sorry if this question has already been answered, I didn't really know what to search for.
Full Version: Why have lossy files a peak over 1.0?
Artifacts due to the mathematics of the lossy encoding indeed might cause a decoded signal to have a peak beyond the full scale. And yes, this indicates clipping. A negative replaygain value might eliminate the clipping, though.
QUOTE (fillip @ Jan 25 2009, 19:25) 
I didn't really know what to search for.
google.com: "peak over 1.0" site:hydrogenaudio.org>>
Replaygain clipping prevention, How does it work?
Additionally, if it's only a little bit over 1.0, it's only a slight error caused by the nature of lossy encoding and it is probably ~0.999 in the original.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.