Topic: Loudness & Peaklevel - Playback vs. Mixdown files (ITU 1770/EBU R128)
(# March, 8th: some corrections after further testing #)
I am sure this is a stupid question, but I can not find the explanation for the following effect:
My RME system (AIO and RayDAT) runs at 176.4kHz. The ITU 1770/EBU R128 meter displays the Analog Output of the AIO and is set to 176.4 kHz.
My audio sequencer runs an arrangement at 176.4kHz, 32bit float. I render the project at 176/32 WAV. I convert the resulting file to 44/16 WAV. When I play back the files in various player applications, both have roughly the same perceived loudness, but both do not achieve the same perceived loudness as the playback from the audio sequencer.
The ITU 1770/EBU R128 meter shows that the peaklevel of the file playback is about 5.0 dB lower than that of the playback from the audio sequencer.
But - playing back both files in another audio application with level meters, both files show peaklevels below -0.1dBFS, while the ITU 1770/EBU R128 meter still shows peaklevels below -5dBFS.
In summary I see two effects here:
1) There is a decrease in perceived loudness of about 5 to 6.5 dB when playing back directly from the audio sequencer (176/32float) versus playing back WAV files from mixdown (either 176/32float or 44/16) in various audio/player applications. (Each playback situation uses the RME AIO Analog Out. In my tests, by changing the playback level in TotalMix by about 5 to 6.5 dB I was able to achieve same perceived loudness for each playback situation.)
2) The ITU 1770/EBU R128 meter in a way reflects this change in perceived loudness as a lower peaklevel. Seen another way - as the level metering of the other audio application indicates - in the world of WAV files 0dBFS seems to be 6 dB lower than the 0dBFS of the ITU 1770/EBU R128 meter (displaying the Analog Out while playing back a WAV file).
I am happy with the ITU 1770/EBU R128 meter as it is. What disturbs me and what makes possibilities for errors in my workflow is, why is the playback of the files about 5 to 6.5 dB lower than the playback from the audio sequencer, even though the same audio device is used with no further gain applied?
I appreciate any explanations or hints. Thank you in advance.