Louder is better.

[IPunchCholla]

We all know that even .5 dB difference in volume will make the louder music sound “better” particularly the bass and treble. Fletcher-Munson tells us why. Yet almost everyone seemed to rant against the loudness wars. i have been making and mixing my music lately and one thing I notice is that louder sounds better (up to a point, but a much higher point than I would have thought) and that DR doesn’t track well with how good it sounds. I have one track that after mixing and using Logic’s mastering assistant sounds best with a LUFS range of 2.9. Of course my music falls into the electronic/noise/punk/metal realm. I can see high amounts of compression being bad for classical, acoustic singer songwriter, etc.

So I have a couple questions: does anyone have any good links to recent research about how humans perceive loudness changes over time? does anyone have any good links to online a/b tests for differing compressions/loudness for various genres of music (I’m on a Mac and foobar doesn’t have the a/b module on Mac last I looked).

 

[egellings]

My impression of loudness wars is that the idea was to compress all the dynamic range out of the music so that radio transmitters could broadcast it at higher average power levels without fear of a loud section in the music causing the transmitter to exceed legal operating limits. That way, transmitter power could be safely increased to get more coverage, which means more advertising income, since the more listeners there are the greater the ad revenue. It's all about da money; always was and always will be. Sorry.

 

[DVDdoug]

I don't think I'd notice a 0.5dB difference. Certainly not from one day to the next. And, that includes the louder one sounding "better".

Of course, the listener has a volume control so you can't really control the loudness.

The main idea of the loudness war is that when your song comes on the radio, or during streaming, it's louder than the previous one. But radio stations use compression to make everything loud. And, streaming services use loudness matching to turn-down loud tracks and turn-up quiet tracks (if they can be turned-up to the loudness target without clipping.

and that DR doesn’t track well with how good it sounds.

That's up to you (assuming you're wearing the producer hat).
I mostly listen to classic rock and I don't always need a LOT of dynamic contrast but I find music that's "constantly loud" boring and I turn it down. Or I just don't listen to it.

i do appreciate "micro dynamics" when drums & cymbals (and other accented notes/parts) "jump out". When music is dynamic and it sounds good, I turn it up! I can have the loudness AND the dynamics.

 

[Blumlein 88]

Loudness increases can be tricky. I recall of a famed mastering fellow detailing how he had increased compression a bit and asked everyone to pick which they liked. Each time the slightly more compressed one was chosen. Even when compression had reached quite high levels. Then he'd let them hear the most compressed version with one having a few levels of lesser compression. Everyone suddenly thought the highly compressed version was a bit yucky.

Usually some compression is liked better. However, one can go stepwise to a point too far. I think that is endemic in recordings today. But that is according to my opinion. So no louder is not better. I will agree there are lots of sophisticated tools to use in increasing loudness levels. Some are very skilled and can push to much louder mixes/masterings that are quite satisfying than is the case with ham-handed compression and limiting only.

 

[Petevid]

yes, the loudness wars weren’t about loudness per se but about compression ie compressing dynamic range and raising it up to the maximum level permitted. You don’t hear about it so much these days but remember when there were lots of complaints about tv commercials being so much louder than the tv programmes. They weren’t, the tv commercials did not exceed the maximum permissible level it was just that their audio was compressed and raised up to around the maximum level so a whisper sounded as loud as a scream, very little dynamic range. The tv companies have since legislated against this.

 

[voodooless]

I recall of a famed mastering fellow detailing how he had increased compression a bit and asked everyone to pick which they liked.

I wonder how the level matching was done? And how much that influenced the results.

 

[Blumlein 88]

I wonder how the level matching was done? And how much that influenced the results.

I don't recall, other than it was commented upon. I'm not sure what is the best method of level matching if there is compression of any significant difference between two tracks. He did use make up gain without which the comparison is pretty goofy.

 

[IPunchCholla]

Thanks for all the replies! I think what I’m getting at is the anecdote @Blumlein 88 mentioned, where even though I am mastering to a constant LUFS, increased compression always ends up sounding better to me in incremental steps. And then it doesn’t. and then I go back to where I started and it doesn’t sound right either,, so I incrementally increase compression… I can A/B my own tracks, but am too close to them. So I was wondering about ways to train myself, perhaps listening to genres outside my own. I think I’m also trying to understand how time effects DR. In that some DR meters will show very low results for a song, but if I look at the DR as measured every half second, it is much higher. So I’m curious about human perception DR…

 

[Neuro]

We all know that even .5 dB difference in volume will make the louder music sound “better” particularly the bass and treble. Fletcher-Munson tells us why. Yet almost everyone seemed to rant against the loudness wars. i have been making and mixing my music lately and one thing I notice is that louder sounds better (up to a point, but a much higher point than I would have thought) and that DR doesn’t track well with how good it sounds. I have one track that after mixing and using Logic’s mastering assistant sounds best with a LUFS range of 2.9. Of course my music falls into the electronic/noise/punk/metal realm. I can see high amounts of compression being bad for classical, acoustic singer songwriter, etc.

So I have a couple questions: does anyone have any good links to recent research about how humans perceive loudness changes over time? does anyone have any good links to online a/b tests for differing compressions/loudness for various genres of music (I’m on a Mac and foobar doesn’t have the a/b module on Mac last I looked).

Higher sound pressure above 85 dB tricks the listener into perceiving it's good. An old trick to sell bad speakers with audible distortion at lower sound volume and measurable even higher distortion at higher sound pressure.
1. Mechanical direct masking of nearby frequencies in the cochlea with and with the help of the outer hair cells of the higher important frequencies above 1000 Hz.
2. Mechanical direct masking of nearby frequencies by the muscles of the middle ear of the frequencies below 1000 Hz.
3. In the brain, nearby prominent frequencies to the direct sound are filtered out via the precedence effect.

All these different "filters" reduce the audible distortion at higher sound pressures.

Compressed music increases the masking effects above.
In other words, perceived distortion is reduced even more at higher compressed sound pressures than at lower volumes.

There is a consensus in the research community that sound and speakers must always be subjectively evaluated with a sound volume below 85 dB, even if the sound is perceived as better at higher sound levels. Toole/Olive uses these levels.

 

[LTig]

I mostly listen to classic rock and I don't always need a LOT of dynamic contrast but I find music that's "constantly loud" boring and I turn it down. Or I just don't listen to it.

Same here. The first time I realized this was with Jethro Tulls J-tull dot com. The first song sounded really great to me but I couldn't listen for more. I think I never listened to the rest of the album.

i do appreciate "micro dynamics" when drums & cymbals (and other accented notes/parts) "jump out". When music is dynamic and it sounds good, I turn it up! I can have the loudness AND the dynamics.

Yep.

 

[DVDdoug]

I recall of a famed mastering fellow detailing how he had increased compression a bit and asked everyone to pick which they liked. Each time the slightly more compressed one was chosen.

He may not have matched the loudness. If you're checking just the sound of the compression you need to take-out the loudness differences.

For your own productions, it would be a good idea to match the LUFS levels to hear what the compression is doing to the music. For your type of music you may want that constant-intensity.

I also suspect that it's a "first impression" thing and with more longer term listening, or normal listening where they have a volume control, people may prefer dynamics.

He did use make up gain without which the comparison is pretty goofy.

Make-up gain is where the loudness comes from. You'd have to back that out unless you just want to confirm that "louder is better".

So I’m curious about human perception DR…

I assume there have been studies but most music producers/engineers use their own judgment. It doesn't hurt to have a known-good reference to "keep your ears calibrated". That goes for everything, not just compression.

I have a suspicion that a couple of generations of listeners, musicians, and everybody else involved is used to loudness war recordings and they probably think that dynamic contrast is BAD! Oh-oh... There's a quiet spot! Keep that intensity going!

 

[Audiofire]

I am mastering to a constant LUFS, increased compression always ends up sounding better to me in incremental steps. And then it doesn’t.

There is a certain volume of air in the room that needs to be filled with acoustic sound. Dynamic range compression is a way to do that and can make details of music easier to hear. This will usually involve speaker interaction with room acoustics, on the other hand the compression can rightfully sound awful with headphones that remove some room acoustics or other distortion. In general.

 

[goat76]

As others have mentioned, compare the files at the same LUFS level. Otherwise, you risk tricking yourself into believing the dynamically limited and louder version is the best sounding just because it sounds louder. However, when you raise the more dynamic version to the same level you will likely find it sounding better, and if you raise the levels even further you will probably notice that the more dynamic version sounds less fatiguing.

But with all that said, you can probably get away with a bit of limiting without noticeably affecting the overall sound quality depending on the type of music. I record and mix my music which is noise rock and I'm starting to think I have to leave my "pure beliefs" in full dynamic range as my mixes are way too quiet, almost to the point I must crank the level up to the limit on my iPhone using my Apple AirPods (possible thanks to the very restricted EU rules when it comes to volume levels).

 

[Audiofire]

yes, the loudness wars weren’t about loudness per se but about compression ie compressing dynamic range and raising it up to the maximum level permitted.

The fallacy of composition is an informal fallacy that arises when one infers that something is true of the whole from the fact that it is true of some part of the whole.

 

[Blumlein 88]

As others have mentioned, compare the files at the same LUFS level. Otherwise, you risk tricking yourself into believing the dynamically limited and louder version is the best sounding just because it sounds louder. However, when you raise the more dynamic version to the same level you will likely find it sounding better, and if you raise the levels even further you will probably notice that the more dynamic version sounds less fatiguing.

But with all that said, you can probably get away with a bit of limiting without noticeably affecting the overall sound quality depending on the type of music. I record and mix my music which is noise rock and I'm starting to think I have to leave my "pure beliefs" in full dynamic range as my mixes are way too quiet, almost to the point I must crank the level up to the limit on my iPhone using my Apple AirPods (possible thanks to the very restricted EU rules when it comes to volume levels).

I don't agree with all you say here. I'd say even with matched LUFS level most people will prefer a moderately compressed over full dynamic range. Then there are other issues depending upon circumstances. Like listening in a relatively quiet car. You will prefer because you will hear more detail a moderate compression to none.

 

[ZolaIII]

Moderate loud mid 70's SPL with equal loudness compensation in more up to date form (ISO 226 2003 or later) is quite fine with me and for front end deticated listening. I usually start a bit lower and increase 1.5 dB every 30~60 min and end in uper 70's, sometimes I do shout out to max no normalisation level of 86~88 dB SPL program but not very often and that's a hard limiter for me.
Edit: forgot to mention that I do EBU R128 on anything I can and then compensate it back +10 dB. Compressed with to little DR (4~6 dB) doesn't sound good, eventually on very low end BT speakers or phone ones it doesn't sound that bad but then again those don't sound good anyway. On the other hand unadulterated TXH mixes with to high DR (22~24) don't sound good on home speakers and not sure even on big movie theatre ones, simply it's too high difference between vocal range and effects.
Mid term from 9 to 12 dB is fine with me and eventually a bit higher on complex instrumental philharmonic peace's but still under 20 dB. ±12 dB is perceived as double louder/quieter.

 

[ZolaIII]

Same here. The first time I realized this was with Jethro Tulls J-tull dot com. The first song sounded really great to me but I couldn't listen for more. I think I never listened to the rest of the album.

Yep.

The Locomotive Breath has a 17 dB DR and costed Ian a lot in the end.

 

[goat76]

I don't agree with all you say here. I'd say even with matched LUFS level most people will prefer a moderately compressed over full dynamic range. Then there are other issues depending upon circumstances. Like listening in a relatively quiet car. You will prefer because you will hear more detail a moderate compression to none.

I think we would need some specific music examples, otherwise we run the risk of talking past each other. Just as an example, we could talk about trying to listen to some classical music in a car with a lot of road noises and we would probably agree with each other that the quiet parts of the music are way too quiet for that particular environment, but in such case, I would have preferred that the compression level was a user setting in the car stereo instead of being applied to the music itself.

If we instead allow ourselves as audio enthusiasts to focus on making music for other audio enthusiasts and put aside outside factors such as noisy environments, I think most of us would agree that a level-matched full dynamic audio production would in most cases sound better than a dynamically limited version.

To be clear, I'm strictly focusing on the dynamic compression/limiting done in mastering with the only intention to make the mix louder, what is done before that to the dynamics in the mix I see as artistic choices or even sometimes necessary compression to make some elements in the mix sounding consistent.

 

[Audiofire]

There's too much dynamic range compression when the volume can't be turned up to decent levels, because continuous power is overloading the ears (on decent hi-fi equipment, not mastered to compensate for loud background noise).

 

[beefkabob]

I can listen to music with different levels of attention, and the less attention I pay, the more pleasant compression is.

I can listen to music in different environments, and the louder the noise floor, the more pleasant compression is.

I can listen to music on different levels of system quality, and the lower the quality, the more pleasant compression is.

I can listen music from different genres, and different tracks within the genre, and what sounds great for one track may sound not so great for another. Classical focuses a lot more on dynamics than most other genres, so it benefits from less compression, unless I'm listening in the car or casually or on a poor system.