This week: I ponder whether blogging is my weak spot or my weakest spot – Can you tell the difference between these two mixes? – My spell-checker doesn’t blink at the word “Californication” – Loudness arms race – Loudness non-proliferation agreement. Everyone is relieved. How did it happen?
I’m either a bad blogger or the worst blogger. I meant to write 500 words about some cool loudness monitoring techniques but then ended up writing a 1,000 word opus and just calling it Part 1 (SPOILER ALERT: It was Part 1 of 3)
We left Part 1 by understanding how a mix becomes louder and we thought everything was going to finally be ok.
Resolution. Catharsis. The Matrix agents run away.
But wait! A well-crafted twist left you wanting more (just humor me).
There was a trade-off. We increased loudness by sacrificing the dynamics. So how do you find the sweet spot? Well, it’s hard. Or at least it used to be hard.
There’s this psychoacoustic phenomenon that’s interesting to observe.
Below you will find two mixes of the same song. There is a slight processing difference between the two mixes. Listen to each one a couple of times and see which version you think sounds better.
Credit: Song written and performed by John Rafferty 2018. Mixed and mastered by Vadim.
Which mix sounds better to you Mix A, or Mix B?
SPOILER: The only difference is that the threshold setting on the Mix A limiter was set to 1db lower - meaning that Mix A would be pushed to be slightly louder than Mix B.
If you don't know this, it just seems like Mix A might sound a little "better" - even if you're not sure why.
We tend to prefer louder mixes. This is largely because our ears/brains kind of have their own little compression “algorithm”. At low volumes, our ears tend to be most sensitive to the frequency range between 2kHz and 6kHz. This is a frequency range containing a lot of critical information for recognizing the human voice, speech, babies crying, etc., In other words, it's a very useful range for us to be sensitive to. At louder volumes, the sensitivity of ours tends to "flatten out" - meaning that we start hearing bass frequencies and high frequencies as being closer in loudness to the range of high sensitivity described above.
In other words, if you listen to a song in your car with the volume turned down very low, you will here the 2kHz-6kHz range (vocals, cymbals) as being significantly louder than the low frequencies (bass guitar, kick, drum, etc.,). As you start to to turn the volume up, the relative loudness of the bass to the vocals starts to even out - and we tend to prefer the latter balance.
If you're interested in learning more about this, search for “Fletcher Munson curve”.
So, because of this effect, people started thinking “hey, if I can get my mix to be louder than the song that was on the radio before it, listeners will think it sounds better”. And so began a period of time during which producers, artists, mastering engineers, etc., began to push mixes louder and louder – trying to outdo each other. The term “loudness war” was coined.
One of the first times that audio professionals started becoming concerned about this phenomenon in public was when the Red Hot Chili Peppers album Californication was released on June 8, 1999.
Admittedly, I actually remember listening to this album when it came out (I was 15) and not really thinking that anything was odd about it. Researching it now, I found that an unmastered version had been leaked shortly after the album's initial release. Below is a comparison of the two versions for the song Californication.
The volumes aren't matched but for me, the clearest differences are in the "space" around the snare hits and the clarity of the guitar riff.
Anyway, moving on...
The difference in loudness between songs on different records used to matter less because you would put on a record, or a cassette, or a CD, adjust the volume on your stereo and the whole album would play comfortably. If you changed records or cassettes, you may have had to readjust the volume but the next album would then sound fine relative to itself.
The big change started in the early 2000s with the advent of the iPod. The iPod made it so you could seamlessly transition from a song on one album to a song on another album without having the processing that radio stations have to level things out. All of a sudden it became annoying to have different loudness levels. You would have to adjust the volume every time a new song came on. iTunes actually had a “volume normalization” function (even in the early days) for this very reason.
This variable loudness problem continued to annoy new listeners as people started moving away from CDs and towards streaming playback services such as Pandora, iTunes, Google Play, YouTube, etc.,
Those people did what annoyed people do best: they complained. A lot.
The streaming companies decided to address this relatively recently and, for perhaps the first time since the Berlin wall came down, pretty much everyone agreed that a positive de-escalation was under way.
What did the streaming services do?
They basically decided to normalize playback volume to a constant level. What this means is that their players would automatically turn down the volume on songs that were "too loud" – giving the listener a more consistent playback level.
There are some differences in how they do this. Some of the services will actually “turn up” quiet songs – while others will not touch quiet songs but only turn down loud songs.
This is a big deal. Some form of streaming accounts for some 47% of revenue in the music industry. This number has been growing in leaps and bounds - a trend that likely won't reverse any time soon.
Engineers and artists have now (nearly) lost the incentive to push tracks louder and louder. In fact, loud tracks would actually be penalized! This is because loud tracks had to sacrifice dynamics to become loud, and they then get turned down anyway. Quieter tracks with better dynamic range will now actually sound better.
So, how loud should your tracks be?
How can you monitor your loudness to make sure you are in the right ballpark?
In the next blog (thank audio god…finally!) I’ll talk about some of the techniques I’ve been using and others that I’ve just started using and which instill a bit of confidence in me when I’m mastering music that I mixed.