@ Bojcha
I hope you know, that
http://radiodak.rs/elma1.mp3
(this is original - before processing with Stereo Tool - correct ?)
already has strong clipping.
I opened it with Adobe Audition 1.5 and played it.
Every other second there is clipping above 0 dB.
But even 0 dB can cause clipping and distortion because of interpolation
between two peaks (though the waveform is normalized at 0 dB).
I am using the TT Dynamic Range Tools (the DR Offline Meter and the DR VST-Plugin):
http://dynamicrange.de
They give the advise in their PDF documentation to normalize at -0.3 dB (as maximum) to avoid distortions caused by the interpolation problem.
Their DR Offline Meter shows a dynamic range value 6 for
http://radiodak.rs/elma1.mp3
So, I would not be surprised (have the clipping in mind) that the extra processing (strong extra dynamic compression) with Stereo Tool results in distortions.
The same thing (distortions) can occur with other music that is also normalized at 0 dB or comes already with obvious clipping (peaks above 0 dB).
There are many (even on original CDs - not MP3 or radio only).
6 is already a high value for the dynamic range (already very limited dynamic range in my ears).
Ironically I am using Stereo Tool the other way round.
Stereo Tool used with the preset "Volume Compression - Normal Soft Compression" widens the dynamic range for many of the radio stations that I listen via DVB-S.
I am using two instances of the TT Dynamic Range VST-Plugin.
One before Stereo Tool and one after Stereo Tool.
For example a station with an original middle dynamic range of 6 comes out (using Stereo Tool "Volume Compression - Normal Soft Compression") with a middle dynamic range of 8.
Here is an example for a MP3 played with Winamp (Apollo 440 - Stop The Rock):
Sorry, all you radio-guys, but I think you want too much (compression -> loudness).
To me this original
http://radiodak.rs/elma1.mp3
- like most of the other files you linked before Bojcha - already sounds distorted (I don't measure - I do listen with my ears).
There are too many high (harsh) tones as if they had cheap monitor speakers with a lack of high frequency response during mastering process.