The Loud Wars why music is being mixed differently

[jckinnick]

XL Recording Studio Guide 2006

Everything Louder Than Everything Else
Have the loudness wars reached their final battle?
By Joe Gross

Monday, October 02, 2006

"You listen to these modern records, they're atrocious, they have sound all over them. There's no definition of nothing, no vocal, no nothing, just like ? static."

? Bob Dylan in Rolling Stone magazine





XL Recording Studio Guide 2006

Everything louder than everything else
Spoon drummer takes a seat at the mixing table
An ode to engineers
The XL recording studio list



The ranting of a cranky old man? Perhaps.

One man's opinion? Hardly.

In August, an open letter from a music industry executive on the state of commercial compact disc mastering and manufacturing was sent to an industry tip sheet/e-mail list run by a music pundit named Bob Lefsetz.

The letter was written by Angelo Montrone, a vice president for A&R (the folks who scout and sign music acts) for One Haven Music, a Sony Music company.

"There's something . . . sinister in audio that is causing our listeners fatigue and even pain while trying to enjoy their favorite music. It has been propagated by A&R departments for the last eight years: The complete abuse of compression in mastering (forced on the mastering engineers against their will and better judgment)."

This compression thing has been a topic of discussion among audiophiles and music fans for nearly a decade. But hearing a music industry executive cop to it was pretty unusual.

The letter was almost immediately reprinted online in audio discussion forums.

"The mistaken belief that a 'super loud' record will sound better and magically turn a song into a hit has caused most major label releases in the past eight years to be an aural assault on the listener," Montrone's letter continued. "Have you ever heard one of those test tones on TV when the station is off the air? Notice how it becomes painfully annoying in a very short time? That's essentially what you do to a song when you super compress it. You eliminate all dynamics."

For those already confused, Montrone was essentially saying that there are millions of copies of CDs being released that are physically exhausting listeners, most of whom probably don't know why their ears and brains are feeling worn out.

He continued, citing an album that proved very popular with Austinites.

"Just to prove that the 'super loud' record has no correlation to actual sales, when we mastered the first Los Lonely Boys record I went to the session and specifically told our mastering engineer NOT to make this a loud record. Could it be that a record that actually had dynamic range could compete? Two and a half million records and a year of constant airplay of 'Heaven' confirmed my suspicion. Loud records are for the birds."

Loud records? Can't you just turn it down? Well, yes and no.

Let's say you go to the store to buy a CD, a brand-new CD of a popular rock band. The group is your favorite, you've been looking forward to this CD for some time. You have the band's other recordings, you've seen them live, perhaps you've even heard the new songs once or twice at a show.

You buy the CD. You take it home and throw it in the CD player. You couldn't be more excited as it starts to play.

But something weird happens as you listen to it. You like the songs, but you don't really want to listen to it for very long and you're not entirely sure why. You take it off. A few minutes, later you put it back on. Same thing happens: You like the music, but you still want to take the CD off. It's more than a little weird.

Condolences. You are officially a casualty of the loudness wars, the ongoing competition among bands, labels and A&R folks to make ever-louder albums.


? ?



Artists, recording engineers and record companies have been trying to make the loudest possible record since the dawn of 78 rpm technology back in the early 20th century.

When 33 1/3 rpm and 45 rpm became the industry standard, engineers strove to make those records as loud as possible as well, often using something called compression during the mastering stage.

Compression means squeezing the dynamic range of an audio signal, usually to boost the perceived volume of a song or performance. Compression works on recorded music the way MSG works on food: It makes everything sound more more. Used with discretion in the recording stage (and even in the mastering stage) it's an invaluable tool for recording engineers.

The idea was the greater the perceived volume of the record, the more attractive the sound would be to the listener. Which meant more attractive to potential DJs, which meant more airplay, more exposure and more sales of the record.

But there were literal physical limitations to this process when vinyl was the primary recording medium ? the music's dynamic range was naturally restricted by the medium itself. During mastering, you could only compress so far; if the sounds were too extreme, the needle would pop out of the groove.

With the advent of compact disc technology in the early 1980s, almost all of this went out the window, as CDs lacked the physical limitations of vinyl.

In theory, this was a good thing. The dynamic range of CDs was far larger than vinyl, and could closer replicate the highs and lows of actual performance. But something else happened.

For the past 10 or so years, artists and record companies have been increasing the overall loudness of pop and rock albums, using ever increasing degrees of compression during mastering, altering the properties of the music being recorded. Quiet sounds and loud sounds are now squashed together, decreasing the recording's dynamic range, raising the average loudness as much as possible.

As Jerry Tubb at Austin's Terra Nova Mastering puts it, "Listening to something that's mastered too hot is like sitting in the front row at the movies. All the images are in your face."

This is why the reissued X album 'Los Angeles' (see story at right) sounds louder at the same volume as the old version, why you turn the 2005 X album down and still hear music, parts that are supposed to be quieter and louder, up front and buried in the mix, at the same time.

For some of you, this difference might be hard to notice at first. Consider yourselves lucky. For some of us, hearing this sort of mastering is like seeing the goblet between two faces in that classic optical illusion ? once you perceive it, you can't unperceive it. Soon, it's all you can see ? or hear.


? ?



Erik Wofford is a producer and mastering engineer in Austin at Cacophony Recorders. He's worked on albums by such local bands as Explosions in the Sky, Zykos and Voxtrot, and finds the loudness wars exhausting to deal with.

"Over-compressing stuff gives everything a flatness," he says. "If loud sounds are the same as quiet sounds, you've destroyed any excitement or natural dynamics that the band creates."

We're sitting with Wofford in Bruce Robison's Premium Recording Service studio, listening to various CDs old and new, running them though the ProTools computer software and looking at their relative loudness. The studio has a woody, '70s vibe. You can totally see Fleetwood Mac recording here (which seems fitting for a man related to the Dixie Chicks). It seems a weirdly inappropriate place to talk about the limitations of modern pop music.

We're looking at the wave forms generated by a number of modern albums. Sound waves should look like what they're called: waves, with sharp peaks and valleys. But the music we're looking at is all peak. It's like looking at a butte or a brick.

"These square waves are a very unnatural occurrence," Wofford says. "It sounds wrong to the ear. You can't hear detail."

There are all sorts of metrics usable to measure loudness, but the Root Mean Squared (RMS) number is a reasonably useful one. It's a measure of average sound level. A smaller RMS number means higher average level; i.e., minus 10 dB RMS is 2 dB louder than minus 12 dB. The maximum RMS value is zero.

Here's the weird part. In the early to late '80s, most pop records averaged around minus 15. (The peak level we see for the old version of "Los Angeles" is minus 14.4 dB RMS.)

Now, modern CDs average at around minus 12 to minus 9 dB. Average.

When a soundwave squares off, something called "clipping" can occur. Clipping in the digital realm means digital distortion, which different CD players handle different ways. Some just won't play that frequency, resulting in loss of dynamic range (you're literally not hearing the whole song). Some digitally distort, which is quite an unpleasant, static-like sound indeed. Some really old CD players skip the song entirely.

There's plenty of clipping on the contemporary songs Wofford and I look at; a red light goes on and stays on the screen when a song clips. Christina Aguilera. Red Hot Chili Peppers. Mastodon. Brick, brick, brick. Clip, clip, clip.

Wofford sighs. "Clipping should just be forbidden," he says. "You used not to be able to turn a redbook CD (the CD from which all others are made) into a manufacturer with clipping on it. That's not true any more."

Thanks to folks on the Internet, there are lists of famously loud CDs. The Red Hot Chili Pepper's 1999 album "Californication" is a notorious example. It clips constantly, and the title track peaks at a whopping minus 5.6 dB, which was really uncomfortable for almost everybody.

That Los Lonely Boys CD Montrone was so proud of? The song "Heaven" averages at around minus 12.5 dB, and peaks at minus 8.9, completely reasonable for modern records.

But the song "Diamonds," on the band's new album "Sacred," clips throughout, averaging at about minus 8.9 dB, peaking at minus 7.7 db RMS.

"I wasn't able to go to that mastering session for the second one," Montrone says from his New York office. "The first record came out when I was with Or Music (the label that released the first Los Lonely Boys album before being acquired by Sony). I wasn't as involved with this new one. I wish I had been."

Who knows if consumers are sick of the band, or the songwriting isn't up to snuff or it has something to do with that louder sound, but "Sacred" thus far has sold about 185,000 copies, and continues to drop on the Billboard albums chart.


? ?



So why aren't more people noticing this sort of thing? One word:

lifestyle.

We listen to music in completely different ways than we did 20 or 30 years ago. For most people, music is listened to on the go, in cars, on headphones while running, on computers at work. Music has to compete with the sound of your car's engine, has to punch through the background noise of street traffic or a loud office.

"Ours is a culture of competition," Wofford says. "Maybe labels think the music has to be super aggressive, super bright, like a kid screaming in a supermarket, to get your attention."

The idea is that louder recordings automatically sound better on low-quality reproduction systems, but this isn't really true in practice. MP3 players such as iPods have their own compressors and limiters, further reducing the dynamic range of recordings, as do computers. A CD doesn't have to be mastered loud; the iPod can make it as loud as everything else it plays.

This is especially true of radio, which, in order to make sure that every song played has a uniform loudness, uses its own compressors and limiters. The idea that a sound has to be mastered loud to be noticed on the radio is just false.

"It's a myth," Tubb says. "Actually, a really loud CD might sound worse on the radio after being fed through a station's processors. (This is what Montrone was talking about with "Heaven.")

This is why the Christina Aguilera song "Ain't No Other Man" (average RMS: about minus 8.4, peak: minus 6.3), which sounds OK-to-irritating on the radio or an iPod, sounds like you are being punched in the face on a real stereo system.


? ?



Yet, bands keep asking for it. That rustling you hear is the mastering community shrugging its shoulders.

"Ours is a service business," Tubb says. "If that's what the client wants, I try to explain the trade-offs in clarity. In reality, we're just trying to accommodate requests from labels or A&R guys or the artists themselves. They'll walk in with a handful of CDs and say, 'I want it to be as loud as this one.' The last five years it's gone absolutely mad."

"Ask any mastering engineer which they prefer," Wofford says, "Something that's super-compressed or not compressed. But they keep their mouths shut about it if they want to keep working."

"It becomes part of (a mastering engineer's) reputation," Montrone says. "Suddenly, you become known for your really loud records. Unless you specify that you don't want it to be loud, they just make it loud. It's become the standard now.

"And it's infected other steps in the chain," Montrone continues.

Mixing engineers often make spec mixes of songs to try and win the bid to mix a particular song or album. "Mixing engineers will turn in spec mixes of tracks that they just slam the heck out of because they think that will get them the gig," Montrone says. "And they're not wrong."

So we're at the chicken-or-egg stage. Is it changing the way we listen to music, or because the way we are listening to music has changed?


? ?



Here's the punch line: The brain can't process sounds that lack a dynamic range for very long. It's an almost subconscious response. This is what Montrone was talking about when he mentioned the TV test tone.

"It's ear fatigue," Tubbs says, "After three songs you take it off. There's no play to give your ears even a few milliseconds of depth and rest."

Alan Bean is a recording/mastering engineer in Harrison, Maine. He's a former professional musician and a doctor of occupational medicine.

"It stinks that this has happened," he says. "Our brains just can't handle hearing high average levels of anything very long, whereas we can stand very loud passages, as long as it is not constant. It's the lack of soft that fatigues the human ear."

This is part of the reason that some people are really fanatical about vinyl. "It's not necessarily that vinyl sounds 'better,' " Bean says. "It's that it's impossible for vinyl to be fatiguing."

And yet, record companies wonder why consumers are buying less of them.

"I definitely think it's a contributing factor," Montrone says. "People have a lot of entertainment options. If listening to music is not a highly enjoyable experience, we're just giving people another reason not to purchase the stuff."

Of course, that's the weird part: Consumers may not know why they are buying fewer CDs or listening to them less or are perfectly happy with low-def MP3s from the Internet.

"That's the big 'too bad' about all this," Bean says: The music is not necessarily at fault.


? ?



The story of popular music is a materialist one ? as playback technology has changed, so has the music.

The LP could hold about 50 minutes of sound (25 minutes a side) if you really squashed the grooves together. As a result, most albums came in at about 40 to 45 minutes. CDs can hold about 80 minutes of sound, and artists have filled them up; the majority of major label pop CDs are an hour or more. The rule seems to be, if you can do it, you should do it.

So it is with mastering: We can make it incredibly loud, so we should make it incredibly loud. Though there is talk in the mastering community of universal mastering standards, it's still just talk.

Again, there is, of course, an element of subjectivity to all this. It is entirely possible that anyone younger than 18 reading this has no idea what we're talking about. They may not bother to buy CDs anymore, such is the availability of MP3s single downloads. To them, popular music has always been hyper-compressed, square-wave stuff, able to punch through background noise with a single snare drum hit, clipping all over the place.

To them, one can say only: You don't know what you're missing.



X: A study in volume vs. loudness


Without getting technical, it's probably important here to define the difference, for our purposes, between "loudness" and "volume." (It's also important to recall that this all gets very relative very fast and that many would argue that there are few true absolutes involved.)

When we talk here about volume, we're talking about the thing which you can control with the knob on your stereo or iPod or boombox.

When we talk here about loudness, we're talking about your perception of a sound at any particular volume.

For example, if you listen to the 1988 CD version of the album "Los Angeles" by the noted roots-punk band X, you have to turn it up to a certain volume to enjoy it. Turn it down low and much of the music vanishes, which is what you might expect when you turn something down.

Now listen to the 2005 CD remaster of the same album. At the same volume as the first version, the songs seem to jump out of the speakers more. The quiet sounds sound almost as loud as the guitar sounds. Turn it down, and you can still hear the quiet sounds almost as well as the louder sounds. This is because the CD has been remastered to bring it more in line with contemporary CDs, which are often mastered louder than ever.

As one employee at a local record store put it, "When we put in older CDs into the CD changer to play in the store, you can't even hear them."

Can't you turn it up?

"Not really," he said. "Because then the newer CDs would be incredibly loud at the new volume. So we don't even play older CDs in the store that often."
? Joe Gross



'If the loudness wars struck the art world'

On the Web site prosoundweb.com, Atlanta rock guitarist Lee Flier imagined a set of remastered masterpieces. Reflecting the sonic damage of pumping up the sound on modern CDs (lost subtleties at the high and low ends as everything gets louder in the middle), the original Mona Lisa and 'American Gothic' paintings become posterized cartoons of themselves. For examples, see Flier's posting at recforums.prosoundweb.com.


jgross@statesman.com; 912-5926



Here is the audio of the interview

http://audio.wnyc.org/soundcheck/soundcheck100206c.mp3

[Mark Alan Miller]

Could it be that the industry is finally waking up to what so many of us have been screaming about all this time???
Wow.

Thanks for that.

[jckinnick]

I sure hope so

[Fletcher]

There is also a general incompetence in both the recording and mixing phases of the music production process. It is far easier to make something sound LOUD than it is to make it breathe.

Far fewer musicians are learning to play the instrument I call recording studio so it has become a game of "how loud can you make it"... distorted vocals sound louder than clear vocals, super compressed vocals are easier to mix because you can just carve out a little space in the guitar assault and the vocals will sit fine from beginning to end... the funny part is that in many ways this "novice production style" was created by the switch to digital audio from analog.

When 16 bit recorders [the Sony 3324 and Mitsubishi X-850] were the norm you had to get your levels as close to 0dbfs in order to employ the full resolution of the bit depth format... these techniques transfered to 24 bit technology where you can actually record farther down in the dynamic range scale and still get reasonable resolution to your audio.

It's funny, I've been transferring some stuff from analog tape into my RADAR V and have found that the average level of what is coming from the analog deck is a full 6db lower than where I usually print the signal when I'm going to digital. On a session last night I backed the levels down so they fit in with some stuff that had originally been recorded analog and was quite amazed at how less fatiguing the sound of the music was... live and learn I suppose.

Peace.

[Judas Jetski]

I'd be remiss if I didn't point out here that I absolutely love X... They are so awesome. I'd been wondering why I didn't like the remastered CDs and why I liked the record albums so much more.

Plus, I think the bonus tracks kill the vibe. There's no need to stick all that extra crap on there. I mean its fun and all, but jeez. Let the album sell itself!

[Professor]

Could it be that the industry is finally waking up to what so many of us have been screaming about all this time???

Sorry.

Couldn't hear you...

radio was up too loud.


OK, I certainly agree that there's room for dynamics, and that music should breathe, etc. so I'll get that out of the way. But there's also some stuff I think wasn't quite covered well in the article, and I'm concerned because I've often had to deal with the backlash of these kinds of articles when they are read by musicians who don't understand what a compressor is or how it works or even what one sounds like when it's running, and yet they get very unhappy with me when I mention that I'm adding a compressor to a track in mixdown. I can't even say the word when I'm dealing with some faculty members - and that's a danger in the other direction because it is an essential tool in the studio.

But I'm getting ahead of myself.
I want to start by blaming radio - which was sorely missed in the articles above.
In the 1980s, there really was a "volume war" going on in the radio industry. Not from city to city, but between multiple stations playing the same formats in the same cities.
Here's the idea which has been proven in many scientific experiments... when given the choice between two identical sounds where one is slightly louder than the other, a listener will perceive the louder sound to be "better" than the softer one.
Stereo salesmen have used this trick for decades. Want to convince your customer that the more expensive amplifier is worth the money, just make sure it is set about 3dB louder than the others when you're comparing them.
Well the radio stations learned the same trick, and one station would simply crank down on their broadcast limiter a little harder to push the overall average level of their signal up a few dB more than the competition. Then, when listeners scanned through the stations, they would hear what they perceived to be both better reception and better audio quality coming from one station when compared to the others, and they would stay on that station.
Well it doesn't take long for the competitor to catch on and up the ante by going another 2dB further, and so on and so on, to the point where the FCC was fining the stations for oversaturation on their bands.
All the while, the musicians were hearing how much louder the songs they loved would sound over the airwaves when compared to their albums, cassettes or even CDs at home. Because it wasn't simply a matter of increasing the volume but of increasing the average as mentioned in the article, so the material really did sound different. Well, so when those musicians showed up in the studios, that was the sound they wanted.

That is such a huge part of the blame that I really do think it deserves at least some kind of mention.

But on the flip side of things, I think it's critical to explain that compression is important for making sure that records are enjoyable to listeners in as wide a variety of listening environments as possible.
It's true, after all, that people do listen on their car stereos or their computers, or with the volume turned way down at their cubicles or on their iPods (or walkmans before them) while on the subway, or in the kitchen while cooking dinner.
Very few people sit down in a quiet and controlled environment to listen to a high-quality stereo system.
I don't know if that is a change or not... and honestly I'm kind of inclined to think that the numbers there haven't changed too much.
So signals need to be compressed to a certain extent so that the music is enjoyable on a wide range of systems to a wide range of listeners.
NO, not to the extreme it has been, but we can't respond with a knee-jerk and make "compression" a dirty word in the studio.

I have to run now, but I'm sure I'll some more ranting later on.

-Jeremy

[Mark Alan Miller]

Yes, I've had to deal with the dirty word "compressor" with some clients, too. They often admit that they're not sure what compression even really does, but they adamantly state they don't want any. Then ask for sounds that are, well, basically achieved with compression...

Good thoughts as always, Professor sir! And yes, the radio volume wars needed to be mentioned, as that started the vicious circle of "make it sound like it does on the radio" which then gets further limited when played on the radio and so on and so on.
In the early/mid 90's there was one artist I worked with who wanted to premaster his record by having it broadcast over the local high-power college station, and then record that off-air to DAT to send to the label... just to exemplify how this stuff can happen.

[Professor]

Oh good, I'm not the only one.
Or perhaps, 'oh bad'.

It's such a frustrating thing because you know they have no idea what they are talking about, and that they're just reacting to some article or something they've heard from other musicians. Sometimmes when they ask what I'm doing while mixing I just explain what the compressor is doing without actually saying the 'c' word, and they seem very intrigued by the concept and happy with the results. Irony.


The radio part of the volume wars is important to remember. And of course the same concept has migrated to television where the morons at the cable stations run about a 15-20dB difference between their programs and the commercials. I'm sure everyone has experienced that little delight.
The vicious cycle comes when musicians hear something but aren't familiar enough with the process to explain what they are hearing. That's why I always laugh when someone shows up at a concert and asks something like what vocal mic a singer is using. Yes the mic is important, but ignoring the entire rest of the signal chain, and the difference between that singer's voice and any other person's voice kind of misses some key elements in the total sound.


Oh well.

-Jeremy

[Mane1234]

Thank you so much for posting this. I was just trying to explain all this to someone last night who will be having his project mastered soon and I was not doing a very good job of it.

[Blade]

I agree with the radio scenerio and everything in general, except this.....

It's not compression that's messing everything up....it's the limiting.

It's all Waves fault!!!! LOL!!!!

All of these brickwall limiters just suck the life out of everything, and I think they blow ass!

I do recording and mixing, and now I'm doing mastering too because I don't want anyone fowling up my mixes.

A CD being loud is not the point, it's the limiting and smashing the bejesus out of it.

I've found a way to get around the L1-L2 madness and still get the disc LOUD!

I SAY.....DEATH TO THE BRICKWALL LIMITER!!!!!!



PIECE OF CRAP!!!!

-Blade

[Knights Who Say Neve]

The article is about digital limiting in the 90's not radio in the 80s. It's the change in technology that made the absurd brickwall limiting possible. Blaming 80s radio is quite a stretch.

And, loudness only make stuff sound better up to a point. That point has obviously been reached and surpassed.

[bobbydj]

Haha.

*takes ring-side seat*

[Mark Alan Miller]

I'd say that blaming radio compression "wars" as a part of the start of the problem is perfectly apropo. The example of a client wanting to "master" through a broadcast chain is a good example of this - it was before digital brick wall limiting was around, and he wanted to be "louder than possible"...

Is 80's radio the whole cause? Heck no. But an interesting part of it's beginnings.

[jeddypoo]

I blame the death of aesthetic and mystery in pop music. I blame "boring" production styles. That actually started in the 70s. With Fleetwood Mac, among others. As nice as those snares sound, that was the beginning of the end to me. Of course, it took until the 90s for that production style to really reach its full fruition. I still have nightmares about the 90s snare sound- a dull, ringing, crystal-clear -wonk- every 2nd beat, with no dynamics at all. I'm thinking Pearl Jam right now.

When I think about it, I spend all this time trying to be a better engineer, when the truth is, I like records that don't really sound very good by modern standards. I read that Dylan article yesterday and sent it to some people. I totally agree with him. There have been a handful of records in the past twenty-five years (I'm 31, BTW) that I would consider having a really great sound to them- The Breeder's "Pod" comes immediately to mind, as do some early REM albums, and pretty much all of Guided By Voices' early 90s albums. Those are recordings with character and personality. Those are recordings with some mystery to them, that imply a different universe.

Now, it seems like people are more concerned with getting things to sound as "real" as possible or something, instead of making something that sounds interesting. And I think that's just as much of an issue as the squash sydrome.

and no, I don't blame digital, because frankly, a lot of the coolest sounding recordings I'm hearing these days are people's shitty Garageband/m-box creations.

[Recycled_Brains]

Now, it seems like people are more concerned with getting things to sound as "real" as possible or something, instead of making something that sounds interesting. And I think that's just as much of an issue as the squash sydrome.

that's a really interesting point. i sometimes see it more as bigger-budget recordings trying to make the music sound better-than-real.... sampled drums, auto-tune, totally clean and clear signals, over-EQing, squashing the dynamics, etc., etc. very uninteresting. everything is so up-front and intrusive. it's like diving face first into a sheet of ice rather than a pool of warm water.

and no, I don't blame digital, because frankly, a lot of the coolest sounding recordings I'm hearing these days are people's shitty Garageband/m-box creations.

same here. 2 of my friends handed me a CD they made on an m-box in their apartment a couple months ago, and it's mind-blowing stuff. it's all in the user.

thanks for posting this article. very interesting read. i esp. liked the stuff about how subconscious the realization of the problem can be.

-Ryan