Lars Lentz Audio™

I have found that I like a certain kind of music from a mastering and quality point of view, and I set out to determine why I like it based on several factors that I think could be relevant.

I used the Youlean Loudness Meter to make the measurements. I took downloaded song files from Bandcamp or ripped from CDs of the music I found best-mastered or best-sounding and measured each.

Here are the factors that I measured:

1. Integrated LUFS (Long-term LUFS)
2. Loudness Range LU
3. Average Dynamics (PLR=”Peak to Long-Term (Integrated) Loudness Ratio”) = (Integrated LUFS – True peak Maximum dB) in LU
4. Dynamics (PSR=”Peak to Short-Term (peak) Loudness Ratio”) = Short-term LUFS – True Peak Maximum dB) in LU
5. Momentary Maximum LUFS
6. Short-Term Maximum LUFS
7. True Peak Maximum dB

What I found was a bit astounding to me:

I thought I would like reasonably loud music – maybe too loud. I expected that the integrated LUFS levels I would like would be in the 9-11 LUFS range. Nope. What really seemed to matter to me were the dynamics of the music. I like the differences between the softest and loudest parts in close contrast to each other.

1. PLR and PSR seem to be the most important to my ears. These characterize the dynamics.
   • PLR was the most important, and somewhere above 8 was common, with 10 to 14 being really very good-sounding to my ears.
   • Interestingly, a PLR that was too high, like above 17, was not so good for me.
   • PSR values of 8 or more were among the best-sounding, with less than 8 being lower quality to my ears. My PSR limit is 8 now per Ian Shepherd
   • The median of PSR values in the 12 to 16 range, was very important to my listening pleasure. This median PSR reading is available on the Sonible product (titled Dynamics there) and not on the Youlean.
2. The true peak maximums were insignificant to me, most were above -1 dB, and some were as high as +3 dB.
   • Boards of Canada’s music were all above +3 dB, but this music was not pleasing to my ears from a mastered standpoint. I like this music very much due to the content, but not the mastering.
   • Most music I like was at 0 or slightly below by 0.1 dB (at -0.1 dB), and I liked the quality.
   • Some were below -2 dB (Tord Gustavson Trio and Matthew Bourne), which made no difference, and at -1, I still liked the music quality.
3. The Integrated LUFS value seemed better if it were around -13 LUFS but higher than -16 LUFS.
   • Classical or ambient piano music was as low as -22 LUFS and as high as -16 LUFS.
     • At the low end of -22 or less, the music was a bit too quiet overall, but if still dynamic by the high PLR value, it was still good to me. This confirmed to me that PLR was important to my ears. It seems that PLR is affected by genre (duh) and so is the integrated LUFS values.
   • Album Integrated LUFS was not measured but should be, and I need to get around to doing that.

In addition to checking the values for music I thought sounded well-mastered, I also checked music that did not sound that great (from a mastering, quality standpoint). I found exactly what I thought – that by checking the not-so-great-sounding music, it had the opposite of what the good-sounding music did.

So this little exercise now drives my own music-making in the following ways that you can find in this post (click here).

AES Guidelines

Here are the AES guidelines I followed, as well as two articles from TapeOP magazine that guided me:

Loudness?

Industry Standard?

Additional Information

In another post, I show you what I do now for my streaming music mastering.