Wednesday, April 16, 2008

Over The Limit

Written by:Rip Rowan
Saturday, August 31, 2002 4:00 PM

I’m a big Rush fan.

Yeah, I know. Me and 50 million other drummers.

I’ve been listening to this band since they showed up on my radar in the late 1970s, and have always followed their tours and new albums. I admit that I fall into the camp of wistful fans who yearn for a return to the art-rock glory days of the band (which pretty much makes me an old burnout) but I still like to hear the new stuff and see what these dudes are up to. And, Rush’s return to a more guitar-oriented (and acoustic-drum-oriented) sound has reignited some of my interest in their performances. Rock music is all about the guitar, and few people are as interesting to listen to as Alex Lifeson. And don’t even get me started about Neil Peart.

The other interest I have in this band is that they have always been at or near the forefront of recording technology. They were one of the first bands to jump into digital recording, where they definitely learned some tough lessons, but the engineering work done on their CDs has usually been top-notch. I can almost always learn something about engineering from listening to the production of a Rush CD.

So you can guess that it was with much anticipation that I awaited my first real listen to the band’s newest CD, “Vapor Trails”. Reviews heralded this album as one of the hardest-rocking Rush albums in some time, with a strong focus on guitars, powerful drumming, excellent bass work, and some of the best songwriting to come from the band in years. And, in listening to the CD, I found all of these things to be true. This is easily my favorite collection of Rush songs in years, maybe decades. It’s incredible work and I earnestly hope it reflects a new and sustainable direction for this great band.

However there was one fact that the reviewers had all left out: this CD sounds like dogshit.

Perhaps you think I’m being a little strong. I think not. This is without prefix or suffix the worst sounding Rush CD ever made. In fact it is so bad that I cannot listen to more than a few songs before I just have to turn it off.

What’s the cause of this sonic catastrophe? There’s no secret here: loudness. Vapor Trails is just the latest CD to fall victim to the current craze of LOUDER IS BETTER production. Rush is not alone. Most of the current crop of rock CDs have been punished by the LOUDER IS BETTER process, and I know I am not alone when I say, once and for all, ENOUGH IS ENOUGH.

Where the damage was done on the Vapor Trails CD is impossible to say for sure. Usually, LOUDER IS BETTER is inflicted by the mastering engineer. It is relatively evident from an investigation of the audio on Vapor Trails that the problems arose in mixing or mastering, not in tracking. However the audio on this disk is so bad that had I been the record label, I would have sued the responsible party for malpractice. Unfortunately, I know all too well that the record label is almost certainly the culprit in this crime, and the band and its fans the victims.

The Label is the Culprit

Record labels have never really understood what makes a record “sound good” and frankly, few even care. Many of the people who sign artists don’t understand their music at all. Instead, they are able to pick up on musical trends, and replicate those trends across the ranks of their artists. Artists that fit into the trend are fed, the rest are starved.

Over the past few years, record labels have increasingly attempted to dictate to the artist and producer the target volume level of the CD. For some reason, record labels have it in their head that “LOUD” equals good, and therefore, “LOUDER” equals better. Not caring to understand even the basics of audio, these morons simply demand more volume (typically from the mastering engineer) and really don’t understand – or care – about the consequences of their demands.

Mastering engineers are caught in a Catch-22. If they do not deliver a product that is appropriately LOUD, then they are consdered inept by the labels and are shunned. If they refuse to destroy the artist’s music, then they aren’t being “team players” and quickly fall out of favor. But if they provide what the customer demands (and remember, the label, not the band, is the customer) then they ruin a perfectly good piece of music, and they know that sooner or later, people are going to figure out why the sound is so horrible, and then the mastering engineer will be blacklisted for having followed orders.

Having said all that I really don’t know what I would do in their shoes. If someone offered YOU the opportunity to master a Rush CD, and then told you that you would have to destroy the sound quality in order to get the job, how would you respond? It isn’t a clear or easy choice.

However what is clear as day is that this CD sounds like dogshit. I cannot say this enough. My God, this thing sounds terrible. It is hands-down the worst sounding CD I own.

Perhaps a brief education about the history of the problem is in order.

A Little History Lesson

Everyone has heard the CD That Is Too Quiet. This is usually your (or your buddy’s) first demo. You pop it in and you can barely hear the music. There are many reasons for the CD That Is Too Quiet, and it isn’t my intention here to go into them all. But we’ve all heard (or made) the CD That Is Too Quiet and regretted it.

Professional engineers, particularly the ones working with digital in the early days of that medium, made some CDs That Were Too Quiet. Usually, these guys had lots of skill and great intent. You get the whole CD laid out in the DAW, and you’ve been careful with your gain structure, and there’s lots of headroom. In one or two places, there’s a freak transient that comes close to 0 dB, but overall the peaks are hitting near –9 or lower, and there’s tons of dynamic range. In general these professional CDs sound pretty good – sometimes excellent - but the average level of the audio is relatively low.

Most older recordings tracked and mixed to analog didn’t suffer these problems. The reason was that traditionally engineers would saturate the analog tape by running it hot, essentally using the tape as a peak limiter at every stage of the process. As a result there are usually no errant peaks in an analog rock recording, and for this reason most rock records are still recorded to analog tape.

The problem with the CD That Is Too Quiet is this: when you put the CD into the CD changer, it’s YOUR music that nobody hears. Well, folks, if you’re a record label exec, that’s the ONE problem that you know just cannot be allowed to stand. Quiet CDs became synonymous with Amateur Recordings, and Loud CDs became synonymous with Professional Recordings.

Understandably, nobody wants to have the quietest CD in the CD changer. Nobody wants to have the one CD that doesn’t get heard. The problem with the LOUDER IS BETTER approach is simply that with any medium – digital or analog – there is only so much signal that will fit in the space provided. Beyond a point, you cannot gain anything without losing something.

Why Be Normal?

The idea behind peak limiting of digital audio started simply enough. Before people got the idea to use a peak limiter on their digital audio, the process of normalizing was used. Normalizing is a strange word that simply means “increase the volume of the signal by whatever amount is needed to bring the highest peak up to 0 dB, full-scale. Normalizing audio during a CD transfer is simply an easy way to get the audio as loud as it can be without changing the dynamics whatsoever. From an audiophile point of view it is the proper technique to get the hottest signal on CD with no distortion of the signal at all.

However, as we’ve discussed, if you have just one transient that jumps out of the signal, then you really can’t get much extra volume out of the signal. Here’s where limiting comes into play: if we just tame the small number of peaks that are eating up the dynamic range of the signal, then we can get the entire signal hotter. Used properly, this results in an imperceptible change to a small number of peaks in the signal and the whole signal can be made louder, sometimes considerably so. This approach achieves the maximum volume while still preserving virtually all of the original signal.

People discovered that with modern limiter technology, you could pretty much ride ALL of the peaks, and squeeze another few dB of gain out of the signal. This approach definitely changes the sonics of the signal because the peaks are being limited throughout the song. However, depending on the source material and your personal taste, this approach to limiting can sound pretty good as long as it is kept in the range of reasonableness. A lot of CDs have been mastered using this approach to limiting, and most of them still sound pretty good.

However, the latest trend is LOUDER IS BETTER. This approach basically ignores any distortion caused by limiting and seeks to make the audio as loud as possible. The idea is to peg the meters and keep them pegged. As a result the signal is just ruined.

WHY IS THE LOUDER IS BETTER APPROACH THE WRONG APPROACH? BECAUSE WHEN ALL OF THE SIGNAL IS AT THE MAXIMUM LEVEL, THEN THERE IS NO WAY FOR THE SIGNAL TO HAVE ANY PUNCH. THE WHOLE THING COMES SCREAMING AT YOU LIKE A MESSAGE IN ALL CAPITAL LETTERS. AS WE ALL KNOW, WHEN YOU TYPE IN ALL CAPITAL LETTERS THERE ARE NO CUES TO HELP THE BRAIN MAKE SENSE OF THE SIGNAL, AND THE MIND TIRES QUICKLY OF TRYING TO PROCESS WHAT IS, BASICALLY, WHITE NOISE. LIKEWISE, A SIGNAL THAT JUST PEGS THE METERS CAUSES THE BRAIN TO REACT AS THOUGH IT IS BEING FED WHITE NOISE. WE SIMPLY FILTER IT OUT AND QUIT TRYING TO PROCESS IT.

I bet you couldn’t even finish reading that paragraph. Get the idea? If louder was really better, then all print media would be printed like the above paragraph.

Editor's note: immediately after printing this article, Senior Editor Bill Park mentioned in a Discussion Forum thread that one point that had not been made in this article was the psychological effect of LOUDER IS BETTER on the listener - that people tend to either turn the music off, tune the music out, or get away from it. Of course, that is precisely the point of the paragraph above in all caps. Bill had actually fallen victim to the point he was trying to make - the all-caps paragraph was so overwhelming to his brain that he either couldn't stand to read the whole thing, or his brain failed to process the message. The irony of this is just amazing, and clearly proves the point that Bill was trying to make: LOUDER IS BETTER means that people actually fail to even hear the music.

Case Study

I went back through some of my collection of Rush CDs to see if my theories held true. What I found was pretty shocking, but not surprising. It turns out that Rush is just a microcosm of what has been going on in music for the last five or so years.

Here is a side-by-side picture showing a sample of audio from five different Rush CDs. On the top is the latest CD, Vapor Trails (2002). Below that, going back a few years, is a sample from the Counterparts CD (1993). Going back a year is a sample from the Roll The Bones CD (1992). Next is 1985’s Power Windows, the first Rush CD to be recorded entirely digitally. On the bottom is a sample from the Grace Under Pressure CD (1984) which immediately preceded Power Windows and was recorded to analog.


A numerical analysis of these tracks offers more insight into the same information:


Peak

Amplitude

# Clipped Samples

RMS (average) Power

Vapor Trails

0 dBFS
110
-9.5 dB

Counterparts

0 dBFS
5
-14 dB

Roll the Bones

0 dBFS
5
-17 dB

Power Windows

-.3 dBFS
0
-18.5 dB

Grace Under Pressure

-.3 dBFS
0
-18.5 dB
a 10-second representative sample was used from each CD in computing these statistics


Power Windows and Grace Under Pressure are classic examples of the Normalization approach to CD transfer (note: I have no earthly idea how any of the CDs were actually mastered, but they still serve as excellent examples of the various approaches to mastering I will discuss). Both Power Windows and Grace Under Pressure show no signs of brickwall peak limiting. Most of the peaks fall well below 0 dBFS, but each one had one peak that hit -.3 dBFS. In the mid-eighties, -.3 dBFS was considered the loudest signal that was safe to put on a CD, since some CD players at the time would treat a 0 dBFS sample as an error. It seems apparent that these two CDs were normalized immediately before transfer to CD master and were not processed with a brickwall limiter.

Roll the Bones shows some signs of peak limiting, but the limiting was a very safe approach. Most of the peaks are just a little below 0 dBFS, but several of them do hit 0 dBFS. However, investigating those samples shows that most of the transient is preserved, leading me to believe that the limiter was used primarily as a safety device, rather than with a deliberate attempt to knock the peaks off of the transients. A little gain was achieved vis-a-vis the older songs: the Roll the Bones sample is 1.5 dB louder than the samples from Power Windows and Grace Under Pressure.

Counterparts is indicitive of an aggressive approach to limiting that still seeks to preserve as much fidelity as possible in the signal. Here it is pretty clear that the limiter was used to shave off a few dB from most of the peaks. Although the analysis software reported that the Counterparts sample and the Roll the Bones sample both had 5 "clipped" samples, investigation of the peaks shows a more consistently limited signal on the Counterparts CD. Not surprisingly, this CD is noticeably louder than the older three samples used in this test. Counterparts is 3 dB louder than Roll the Bones and 4.5 dB louder than Power Windows and Grace Under Pressure. As you probably know, that means that you will perceive Counterparts to be twice as loud as those older CDs. And, while the trained ear will recognize the sound of Counterparts as having been processed with a limiter, the sound is still relatively open and more-or-less unmolested. Audiophiles will probably disagree with me here, but in the world of rock music, a little bit of peak limiting doesn't necessarily ruin the sound.

But what is going on with Vapor Trails? The numbers quickly report the truth we were hearing with our ears. The average volume is a whopping 4.5 dB louder than Counterparts. But where Counterparts was able to achieve a 4.5 dB volume boost with almost no increase in audible distortion, in the Vapor Trails sample almost every peak is clipped, many of them severely so. And, as we shall see, the limiting is so severe that the songs have no punch, and just slam out of the speakers like a loud blast of white noise.

What Price Volume?

Have a look at this and tell me what you see:


One of thousands of ruined kick drum transients from the Vapor Trails CD


What you are looking at is a serious square wave. Note in the top swing of the clipped wave how the high-frequency harmonics have had all of their peaks shaved off? This is why clipping sounds harsh. Now note how in the lower portion of that wave the bottom is just perfectly squared off? That's a pulse of white noise. Now note how all of the peaks in this signal are all hitting the top of the scale? This is why the CD has no punch. A powerful blast of the kick drum should be louder than the rest of the signal and should have a sharp transient on the front of the wave. In the case of this song, however, it just produces a quick blast of distortion and a dull thudding sensation.


If there was one or two places on the CD that had been tortured like this, it would be one thing. But that's not the case. Every peak on the whole album is wrecked. In fact I didn't have to try to find an ugly picture to illustrate the damage: the question was "which one do I choose?"

You're probably thinking, "You've got to be kidding!" But I'm not. These are just a few of the thousands of examples of the trashed audio on the Vapor Trails CD.

Now, anybody with any experience in audio should be able to look at these examples and immediately know, "That won't sound good." And, it doesn't. So the question is, why is this CD wrecked, and by whom?

Notice how, when the signal clips, the whole signal is being destroyed? If this had been caused by an error during tracking (say, the kick drum track was too hot) then there would still be detail left in the signal. Therefore, we know this distortion was caused either during mixdown or during mastering. I'd be willing to bet that it was caused during the mastering process. At least I hope it was during the mastering process. At least that way there's some chance that one day I'll be able to buy a remastered copy of Vapor Trails that's worth listening to.

But before you think I'm accusing the mastering engineer of incompetence, remember my earlier discussion. I doubt very seriously if any engineer actually wanted to do this to the music. No, only the decision of a record label executive could destroy music like this.

What a shame. What a crying shame.

Solutions

I can sit here and rant all day long. The real question is: what can be done about it?

The good news is that LOUDER IS BETTER is definitely a self-correcting problem. Because this stuff just plain sounds bad, and sooner or later (hopefully sooner) people are going to realize that the music doesn't "rock more" or "cut through better" but that it's just plain annoying.

Because the simple truth is that audio such as this does NOT cut through better. In fact, in all probability this song will be QUIETER on the radio than, say, Roll the Bones! How, you may ask, could this be possible?

The answer lies in the simple fact that the radio station uses compressors and limiters as well. The station's signal processors are also designed to get the hottest signal on the air. As such, they expect a certain amount of peaks in the signal. A broadcast processor that can't "see" any peaks is simply going to clamp down on the whole signal. In the end, the song is no louder (and maybe quieter) than other, more dynamic material - AND it is further penalized because it has no punch and is very harsh.

So to you record executives who think you have identified the magic way to ensure that your song is louder on the radio, think again. It just isn't so.

And as far as being the loudest CD in the CD changer, has it ever occurred to anyone just exactly how annoying that is? Let me tell you, you won't catch me loading Vapor Trails into MY changer with a bunch of older Rush CDs. It sticks out like a sore, bleeding thumb.

As I have said before, Rush is not the only band to fall victim to LOUDER IS BETTER. As a matter of fact I have had some of the CDs which I have engineered fall victim to LOUDER IS BETTER. I don't mean to pick on Rush, one of my favorite bands of all time. However, let's look at some facts.

More than most other bands, Rush has proved itself capable of consistently delivering music that pleases the fans. Rush's success has not come through a string of chart-topping hits, clever marketing, or sex appeal. Rush's success comes from its close alliance with its fans.

I wonder what would happen if Rush fans complained about the sound of this record? To my mind, the only thing that the record company mooks will pay attention to is audience disapproval. So I encourage you to write to your favorite bands and tell them that you will quit buying their CDs if they insist on trying to make them the loudest CDs you own.

Oh, and when you write them, WRITE IN ALL CAPS.


Editor's Note: a dumbed-down version of this article was published in the January 2004 issue of Wired magazine. You can read that version here. There was also significant discussion of this topic on Slashdot.

Tags:

52 comments so far...

Re: Over the Limit

hey! I'm glad to see ProRec is back online. This article is a classic. When will there be more new articles?

By dave on Saturday, July 28, 2007 8:12 PM

Re: Over the Limit

Yeah.. we missed you. Still reference your Delta 1010 article.

Are you looking for contributing editors?

By liberoj on Sunday, July 29, 2007 1:18 PM

I'm glad it wasn't just me

I am not an audio professional (in fact I've made my own CD That Is Too Quiet) but I had a similar experience with Vapor Trails. I found it unlistenable. I'm glad that someone who actually knows what they're talking about had the exact same experience.

By slar on Tuesday, July 31, 2007 11:27 AM

Re: Over the Limit

Whew, glad your site is back.
Still can't listen to VT for a great length of time. Even tried limiting the tracks with Audacity, but like Rip says, that doesn't restore any sonic goodness. There's STILL apparently no news on a remaster.

By Mike on Thursday, August 02, 2007 2:07 PM

Re: Over the Limit

Another person glad to see ProRec back. Rip, you are right on: Vapor Trails sucks in terms of audio quality. This is a classic article, a must read for anyone recording anything.

By Ken De Vries on Thursday, August 09, 2007 12:18 PM

Re: Over the Limit

And apparently Snakes and Arrows isn't much better. The drums are so squashed that there is no punch and definition at all. The live versions of these songs are 1000 times better and much more appealing. I wish the audio trend would stop pandering to the least common denominator. Super compressed, over limited tracks designed for playback on iPod earbuds is definitely NOT the way to go for the future of music. By the way, welcome back! We missed you.

By imagimotion on Thursday, August 16, 2007 1:03 PM

Re: Over the Limit

Could not possibly agree more!! Unfortunately, if the tools are there - they'll be misused and abused.

By goldivox on Thursday, August 16, 2007 9:15 PM

Re: Over the Limit

Man this has driven me nutz for years!! I'm constantly showing my clients comparisons of the smashed cd references they bring and well recorded and unlimited mixes at equal volume in the studio and the look of shock in their faces when they hear what hard limiting does to music is priceless! This is so important for the future of music, I'm so glad that more and more people are rasing a stink about it !! Keep up the great work and welcome back, this site was sorely missed!

By Leo Alvarez on Saturday, August 18, 2007 11:32 AM

Re: Over the Limit

One word. Amen!

By MICHAEL on Thursday, August 23, 2007 10:19 AM

Re: Over the Limit

Hey, I don't believe you mentioned the analysis software you used to get the power results. Could you tell us what it was?

By Matt on Friday, August 24, 2007 5:00 AM

Re: Over the Limit

hey nice article thanks a lot

By leakeg on Thursday, August 30, 2007 4:56 AM

Re: Over the Limit

It is a nice article. I do have to disagree with some things. Showing images of the ruined kick sound wave doesn't tell me how it actually sounds. Actually, judging by the latest CD screenshot sample, it's not really that squashed. I can't believe it sounds that bad, and probably it's not as loud as a lot of hard rock CDs out there.
The other thing you forgot to mention, which is of vital importance, is that this squashing is usually requested by the band and approved by the band. Specially in a band with so many albums, the label does what the band requests, engineering wise. They're not victims, it's their artistic choice to squash it a little to make it louder. Eventually with time it is becoming a desirable characteristic in a rock record. In the early days of guitar distortion, it was considered an abomination and audiophiles argued that the human ear finds all distortion unpleasant.

By Jonathan Grand on Thursday, August 30, 2007 1:11 PM

Re: Over the Limit

Also, even compression is a "new thing". A lot of old time engineers still hate compression and think it's absurd to modify the original transients of a sound recording, but when they hear a Chris Lord-Alge mix they all ask "wow... amazing... how did he do that?"... So it's funny that you, Rip, an engineer that I respect and admire so much, can't see the irony behind all this anti-loud-mastering fad.

By Jonathan Grand on Thursday, August 30, 2007 1:15 PM

Re: Over the Limit

Jonathan, (1) the kick drum is a square wave, so you can't say it "isn't really that squashed," (2) it DOES sound that bad, (3) it is within 1 dB of being the loudest rock CD I've been able to discover, it's really really hard to make a rock CD that averages above -9 dB that isn't just plain white noise, (4) the band (and / or label) usually say that they want it LOUD, not "limited-sounding", otherwise they'd have limited the snot out of it in tracking, (5) I'm pretty damn far from being some kind of esoteric anti-distortion snob. But, more on all of this in my next article....

By editor on Thursday, August 30, 2007 3:47 PM

Re: Over the Limit

I mean the part where you posted images of the peaks in all of their albums? Most Limp Bizkit and Linkin Park CDs sound amazing and you can't even see peaks if you dont zoom in, just a solid block! :) Of course, probably those cd's are way better mastered than Rush's latest cd, but the point is, "visually" it's not that squashed... So you think Rush were disappointed when they heard the mastered CD? It would be interesting to know the ins and outs of this process as a whole.

By Jonathan Grand on Thursday, August 30, 2007 5:53 PM

Re: Over the Limit

"I mean the part where you posted images of the peaks in all of their albums? Most Limp Bizkit and Linkin Park CDs sound amazing and you can't even see peaks if you dont zoom in, just a solid block!"

Snerk.

By Allen on Thursday, September 06, 2007 8:18 PM

Re: Over the Limit

My point is, you have to listen to it and how it sounds. Relying only on your eyes is like mixing looking at VU meters.

By Jonathan Grand on Friday, September 07, 2007 2:11 PM

Re: Over the Limit

Allen: being an audio snob and assuming things arrogantly is not a very professional attitude. Assuming you want to be taken seriously.

By Jonathan Grand on Friday, September 07, 2007 2:14 PM

Re: Over the Limit

I recently bought the new Mae album and, although I'm a huge fan of their previous work, I just couldn't listen to the new CD more than once. There aren't any dynamics whatsoever, and every sound is in your face. Constantly. It's just incredibly fatiguing.

I hope to see the day when a bunch of influential mastering engineers start talking to each other and collectively tell the labels to shove their Square Wave Philosophy where the lights don't shine.

By Alex on Monday, September 10, 2007 8:02 PM

Re: Over the Limit

...Most Limp Bizkit and Linkin Park CDs sound amazing...
Very disputable statement, whether or not one is an "audio snob"... But anyway it is absolutely different style of music. Maybe we should remaster The Beatles to sound like Linking Park? And even then they won't be sounding like Linkin Park. They will be just sounding like shit. Because in case of Linkin Park it's not just the mastering issue. That "solid block" was intended from the beginning of recording process, starting with tracking, and maybe even composing and arrangement of songs.
It maybe good or not, but it's definitely not what we want to hear from Rush (or Rolling Stones... or RHCP... or Ozzy Osborne... - just a few examples that I couldn't play to the end even in iPod).

By Alex_K on Thursday, September 13, 2007 1:48 AM

Re: Over the Limit

I tend to agree with Alex_K (above). Material that sounds good when it's heavily limited is usually created to sound like that from the outset. Just cranking up the limiting on an otherwise dynamic recording doesn't achieve the same result.

By editor on Thursday, September 13, 2007 8:42 AM

Re: Over the Limit

Hmm ok. Interesting. Hey Rip did you get my message? I sent it directly to ProRec.

By Jonathan Grand on Thursday, September 13, 2007 9:12 AM

Re: Over the Limit

I'm no musician, but anyone can notice this if they pay attention: listen to an 80s rock CD at a comfortable volume, the drums will usually be louder than the vocals. But put on a new record and set the volume so the vocals are about the same level. Attention to the chorus, when they go the loudest. Where are the drums? They're little more than clicks, barely audible in the wall of sound. They turned the drums into fuckin' BACKGROUND NOISE!

By An amateur on Sunday, September 16, 2007 9:12 PM

Re: Over the Limit

...Where are the drums???... That's it! We suppose that rock music is all about the beat. But in today's records it's all about loud stupid SHOUTING, no drums, no beat, no rhythm. Funny thing, they do it even with dance music! Recent example. I bought Madonna's Confessions on a Dancefloor. At first it seems that everything's in place: lows, highs etc. I try to turn the volume up, because I like when the drums are slightly "shaking the air" on dance music. What the hell? No "airshaking", just unbearable shouting! I turn the volume down and never crank it up again.

By Alex_K on Monday, September 24, 2007 8:23 AM

Re: Over the Limit

I completely agree with the article; what can we do though? I remember buying vapor trails and within a week I sold it. I couldn't stand listening to it...there latest is almost the same too-it sounds horrible, I fear this is the end of high quality music(if CD is even that). That is until I bought the new Foo Fighters. Wow what an audiophile recording-seriously. I'm not a huge fan I do not own any of their older music but this new CD has dynamic range a plenty. Anyone into Megadeth or The Tragically Hip? Their latest releases sound awful too. It makes me sad because their earlier stuff is decent. Who do we email or complain to? Another dissapointment is the new Matthew Good actually the last 2 albums the sound quality is brutal...

By Jay on Tuesday, October 02, 2007 6:21 PM

Re: Over the Limit

Great to see Prorec back on line. This article was one of the first I read regarding dynamic range. It turned me on to this loudnes war thing. Another article "What Happened To Dynamic Range," is another excellent read. Written by a mastering engineer, no less. Maybe things are looking up! http://www.cdmasteringservices.com/dynamicrange.htm

By Bjorn on Wednesday, October 03, 2007 11:35 PM

Re: Over the Limit

Keep up the educating. There can be no loud without 'less loud'.
Check out the latest Beastie Boys album - plenty loud enough without being smashed, circa -14dB rms.

Make it sound great. Turn it up.
- Mastering Engineer

By Adam on Monday, October 15, 2007 4:21 PM

Re: Over the Limit

Gosh, So true, I can't listen to any of this new produced shit, because it sounds like a wall of DUNG being hurled out of the speakers. I realize that this covers up most of these young no-talents lack of ability but there are some good artist putting out the same barage of masive puke. I hope we get back to quality sound and recording and leave this bastarding traffic jam behind.

By musician on Wednesday, October 31, 2007 8:07 AM

Re: Over the Limit

OMG, could this be why my ears hurt when I'm experimenting with Louderbox? It's cool and all... but I think I see^H^H^Hhear precisely what you mean. Thanks for this.

By smashed on Wednesday, October 31, 2007 9:14 AM

Re: Over the Limit

You are the man.

By Sir Migs on Monday, November 19, 2007 10:02 AM

Re: Over the Limit

nice article, but i think you're over simplifying the blame aspect. I'm currently doing a study into the loudness wars is that more often than not it's actually the band themselves that want it.
Some mastering engineers do it, because it is their style. I find labels only intervene when the subject is a mainstream act (whether they be pop, rock, hip-hop or something else)
Unfortunately I don't see an end to this any time soon

- Leigh

By Leigh O'Gorman on Wednesday, December 12, 2007 10:41 AM

Re: Over the Limit

Snakes and Arrows doesn't sound TOO bad... I can actually enjoy that album! Vapour Trails... First song is just barely listenable... Second song, I turn off the album. It REALLY annoys me since it'd be one of my favourite albums if I could listen to it.

I recently showed my dad a song from Vapor Trails and then showed him a song from the vinyl rip of Stadium Arcadium (which has NO compression applied to it at all) which I have. He was amazed at how bad it sounded. I even put on Slayer's Reign in Blood for a few minutes and was like "See, even thrash/death metal USED to sound good!"
I was listening to some stuff at other work the other day that was sourced from vinyl and then I REALLY fancied listening to Muse... Could I? Just barely; Absolution's compression is really noticable after listening to good-sounding albums!

There are some good modern CDs though. Marillion albums aren't too compressed and sound very good. Tool's 10,000 Days is a good example of a modern sounding album that has a moderate amount of compression and still has decent dynamics. Unfortunately my favourite band (Dream Theater) has been dynamic range compressioning to crap for the last few albums! Why I go through random phases of listening to tons of live stuff -- usually hasn't been dynamic range compressed as much...

Another one I found interesting was listening to a VERY rare version of Metallica's Master of Puppets which had been remastered for a DCC Golddisk in the 90s. It does sound a lot better (you can actually hear the bass), but what do I find? Dynamic range compression! They dynamic range compressed a limited edition, audiophile-only edition of an album! WHY?!?

Oh well... It's still better than no music... Just... *rolls eyes*

By SirCanealot on Wednesday, December 19, 2007 9:22 AM

Re: Over the Limit

I AM a radio engineer for a classic rock station. Even going to CD was a detriment to good sound. I've done A/B comparisons of cd's and vinyl, and there is no comparison. I'll admit, analog requires good equipment and upkeep in order to sound good, and that is what most people liked about CD. It's consistent, and brighter than most people could expect from their cheap analog systems. But that CD brightness, though consistent, is in large part manufactured by CD's sample rate's inability to accurately reproduce anything above about 4 Khz. After that, it begins to distort, and many of the distortion artifacts manifest themselves as higher frequencies. Where at the same time, high frequencies are getting distorted up to and including complete disappearance, and almost never what they were recorded as. Allot of that brightness is high frequecies created from a combination of midrange and high frequency digital distortion. That's how the subtlties and smoothness and realism dissappear.
Anyway, it would require a sample rate of at least 5 times that of CD, as well as greater amplitude resolution, in order to rival the sound of analog - fortunately, that is just about what DVD audio is capable of.
Now, in mastering, if it is done in the digital domain, every step of the way, any change whatsoever will add digital distortion, not unlike the distortion which makes a CD bright.
An experiance I had one time on the way to do some night work at the station.......blew my mind. I kicked on the tunes in my van, and it was Black Sabbath "Sweet Leaf" so I turned it up. But sonically it sounded so absolutely incredible, I soon began wondering what source I was listening to. Upon first check, it appeared I was listening to the radio, to my own station. But that couldn't be. My station was winning the loudness war, and I generally couldn't stand listening to it. So, just incase, I punched around the dial real quick and came back. Sure enough, not only was it definitely my station, and not only were we the loudest thing on the dial, but we were louder than usual by what seemed several orders of magnitude, and we were by far the cleanest sounding thing out there. I was in total and estatic awe....and with no understandable expalnation in mind, fully enjoyed listening to the rest of that song. I mean, it like filled the entire van from corner to corner.....just deliciously (I could spend paragraphs describing all the differences). Well. when I got to the station, I pretty near ran to the FM studio to quiz the jock on what he'd just played, and what he did differently, or whatever. As you probably have guessed, all he did was slap a piece of vinyl on the the turntable.
It was then I realized how we have screwed ourselves. Digital sucks.....especially though compressors.
But this story doesn't end there. It turns out, that the actual piece of vinyl he was playing, was actually mine. I had forgotten that I lent it to him some time back. And so I know a little about the history of that particular slab. Infact, I was at least the third owner. It had no sleave. The previous owner didn't know anything about care, and generally didn't even put it back in it's jacket. It just sat out in a pile of unjacketed vinyl. AND, it was so dirty and scratched, it was grey.

And yet, on-air, it blew the doors off of anything we played off of CD....hands down....no questions asked.

The moral is, audio compressors act upon digitally derived distortion, entirely differently than they act upon analog distortion. This could be why various metering and other tests and indications may be worthless. If the signal you are starting with contains digitally derived trash, then the compressor's algorythm of reading and deciding how to act upon it, is totally erronious.

Thus, if in original recording, the individual tracks are recorded in digital, all subsequent changes, be it a minute volume change, or a tweek in the tone, as well as going though any compressor/enhanser, to say nthing about the conversion back into analog, will all result in manufactured artifacts - none of which are music. And thus, it is probably not the compession itself, and probably not the shaving off of even a great number of peaks, which causes the horrendous sound we hear on VT. It's more likely digitally distortion being improperly acted upon.

The bad news is - nothing can be done about it in the compressors, because these artifacts have an infinite number of possible departures from natural. And so no compressor could be built with the expectation of handling them.in stride even if knowing that they're coming.

The solution is to use a suficiently resolute sampling scheme starting at the original recording, and continuing unbroken through the system. Ultimately acting upon that signal within one piece of equipment, making all changes to the signal in one step. And then setting it out for public consumption at a minimum of DVD quality. Then, massive compression will sound killer.

By Kentech on Saturday, December 29, 2007 5:16 AM

Re: Over the Limit

I agreee. Vapor Trails is like a huge cinder block wall. I think the songs and playing are great but the "sound" of the album is just too much to injest. Great article that explains the issue with "sound" evidence.
Regards,
Chuck Bruce
Member of the Rush Tablature Project

By Charles D. Bruce on Saturday, December 29, 2007 4:17 PM

Re: Over the Limit

Kentech's comment is full of anti-digital and anti-CD misinformation, but Rip's article is spot on!

The most annoying thing about this whole issue is that no obvious solution comes to mind, at least not one that the record industry has shown any inclination to adopt. The best suggestion I have seen is that the labels release an uncompressed audiophile master a couple of months after the LOUD commercial release, with a bonus track or two so that hard core fans buy both versions: now that might make commercial sense!

By Arg on Sunday, December 30, 2007 10:46 PM

Re: Over the Limit

Dynamic range compression isn't limited to rock music. Case in point: I own a copy of Frederic Chopin's two Piano Concerti conducted by pianist Krystian Zimerman and the Polish Festival Orchestra (Deutsche Grammophon 459684, released November 1999). Upon listening, I could tell that it was noticeably "louder" than similar classical content - so I performed some analysis, shown below (I used the open-source "normalize" utility):

-8.3808dBFS Chopin Piano Concerto No. 1, first movement

-13.5691dBFS Beethoven Piano Concerto No. 2, first movement (Kissin/Previn, Sony Classical)

-12.3891dBFS Rachmaninoff Piano Concerto 1, 1st mov. (Ashkenazy/Previn, Decca)

-12.2954dBFS Rachmaninoff Piano Concerto 2, 1st mov. (Rubenstein/Reiner, RCA recorded in 1956!)

It's plainly obvious that the DG engineer(s) decided to extend the loudness war to classical music, at least with this particular title. Or, was it just Zimerman? I don't know. Other DG titles I own have volume levels consistent with other labels.

By P.Read on Sunday, December 30, 2007 11:35 PM

Re: Over the Limit

... If someone offered YOU the opportunity to master a Rush CD, and then told you that you would have to destroy the sound quality in order to get the job, how would you respond? It isn’t a clear or easy choice...
Actually, yes, it IS a clear choice for me. I would do right by both the artist and my own COMMON SENSE. If the record company doesn't like it, get someone else. I SELL OUT TO NO ONE!! "Nuff said.

By mstcraig on Tuesday, January 01, 2008 9:16 PM

Re: Over the Limit

I do alot of live recording in Austin TX. In transferring my recordings and mastering my recordings, I use absolutely no compression, and the output level is set to the highest sound peak in the whole recording, which I make sure does not exceed 0db. Funny, but I have to tell people who listen to my recordings and the bands that they need to listen to them on a home stereo, not in their car or thru an ipod or headphones, as it will sound like shit. They are perplexed until they come to my house and listen to it through my stereo, which they respond, it sounds like we are actually in the room playing. I have to tell this to friends of mine who have all there music on mp3s, that it is crap and they are listening to crap. Funny, but this is all the result of commercialization and mass marketing which obviously has turned this country into a bunch of 'follow the leader' zombies. Sad, but at least some of us know the real truth.

By David on Tuesday, January 01, 2008 9:42 PM

Re: Over the Limit

... If someone offered YOU the opportunity to master a Rush CD, and then told you that you would have to destroy the sound quality in order to get the job, how would you respond? It isn’t a clear or easy choice...
Actually, yes, it IS a clear choice for me. I would do right by both the artist and my own COMMON SENSE. If the record company doesn't like it, get someone else. I SELL OUT TO NO ONE!! "Nuff said.

By mstcraig on Tuesday, January 01, 2008 9:44 PM

Re: Over the Limit

Reading the biography of RUSH, during the mastering process of Vapor Trails, the band and producer heard a distortion while the engineer was 'working' on the audio. They were unhappy with it and halted the process to go back to the studio to find the source of the distortion (digital clipping). Of course they couldn't find it - it was the mastering engineer they had employed making the audio distort. On return to the mastering suite, hey presto - it's back again, while Mr Weinberg is squashing the crap out of their songs.
I recently sent a project to Howie Weinberg to get mastered, and it came back full of digital clipping - absolutely unlistenable in my opinion, but the client was pleased. I explained how awful it was and ran it through my DAW's 'declipper' to remove some of the trash - it straight away sounded easier to listen to, but still pretty bad.
If you look at a sample of Mr Weinbergs work, you'll find that is his trademark. Clipped full code samples that completely wipe out the dynamics of the track. I haven't heard one decent thing that he's done.
My 2cents worth.

By Tim Julian on Wednesday, January 02, 2008 1:22 AM

Re: Over the Limit

excellent article.

also check out turnmeup.org
and "loudness war" on wikipedia.

By rekall on Thursday, January 03, 2008 12:24 PM

Re: Over the Limit

Well folks....I don't see this problem going away. Anyone who cares about quality in commercially available music is...unfortunately.....the minority. Convenience formats have taken over and that trend started way back with the 8 trk cartridge. Then cassette, minidisc, etc. CD came along and by the time technology improved enough to make it a viable alternative to quality analog formats the dumbing down started. Its a consumer product after all and demand led by the "average" music consumer will dictate the trends. iPod toting kids don't want to have to listen to anything...they want to have it *thrown* at them. I agree that the era of high fidelity is long gone....never to return. Any new format aimed at actually improving available audio quality will fail miserably. Why? Very few care anymore. And if a format doesn't sell well, it goes away. The "average" listener (victim) can't tell the difference because they've likely never heard much thats of good enough quality to show the contrast. The plastic gets prettier and the cheese gets stinkier. My convenience format? Vinyl transferrs via MC cartridge to a 1/4" half track Revox A77. Whatta sound....

By Gtr_tech on Monday, January 07, 2008 1:09 PM

Re: Over the Limit

Great article and comments from all. Same thing I heard when a fellow Rush fan gave me VT and it blew me away far away - I understood why he could had it off that easily - I haven't heard it since... by the way I have recorded the CD no one could hear and even our pro recording is actually listenable because it was mixed "old school" and some people have told us it should be remastered, which we as a band have fought tooth and nail. Thankfully we all agree it sounds better than most things out there - you just have to LISTEN, we are not going to shout at you (unless that is for the moment it is dynamically appropriate. Check us out and tell us your opinion www.pagasys.com of our album "Now You're Here" and kudos to our engineer Pat Sample of Paradise Sound Recording.

By Jian-rockgtr on Monday, January 07, 2008 7:03 PM

Re: Over the Limit

Great article! I'm not a pro recorder - just a lover of music. I've wondered why over the last decade or so I've found it so hard to find an album that I could listen to all the way through. I think I have my answer. I too am a HUGE Rush fan, but haven't loved many of their last few albums. I thought it was the music, but I think it was more of the production after reading this. It's funny, because a few of the bands/albums that I can listen to frequently have been mentioned here as examples of "good" recording (Tool, Grace Under Pressure, Power Windows...). Maybe I have a better ear than I thought!

Keep fighting the good fight! Audiophiles and music lovers alike deserve better!

By musicfan on Tuesday, January 08, 2008 1:21 PM

Re: Over the Limit

What's ironic is that I find the music on Vapor Trails to be outstanding. It's one of my favorite Rush albums. Their latest release "Snakes and Arrows" is sonically beautiful, but the songs are sub-standard. If only we could take the production from S&A and combine it with VT. That would be one killer record.

By imagemagician on Tuesday, January 08, 2008 5:44 PM

Re: Over the Limit

Great article.
I own all of Rush´s albums...I only listened to Vapor Trails once. Is it any good?
I can´t tolerate the sound, so I can´t play it again....
Test for Echo was also too loud.
Hope we get a proper remaster one day

Thanks

By Javier on Friday, January 18, 2008 11:30 AM

Re: Over the Limit

I finally managed to rip the 96 kHz LPCM audio from Snakes and Arrows "MVI DVD" using DVD Decrypter and VLC wizard (uncompressed integer mode does the trick). I had high hopes. . . Well, some of the tracks are so clipped that the waveform has almost no transients. I could have done a better mastering job. Of course, that isn't saying much. . . I must say I am enjoying the album very much but I wish I could crank it without my ears hurting. I am disappointed that my heroes apparently are shoddy engineers.

By Mike on Wednesday, January 23, 2008 11:56 PM

Re: Over the Limit

There's a second opinion on the subject at
http://www.moozek.com/2008/02/03/the-loudness-war-oh-please/

Check it out, comment and subscribe! :)

By Jonathan Grand on Wednesday, February 06, 2008 1:35 AM

Re: Over the Limit

Thank You!!! Rip outstanding job, and very cool using the Wave Form grahics to illustrate this horrible trend. I own a recording studio and although I'm not a mastering engineer, most of my clients skip professional mastering and I end up reluctantly mastering it myself. I say reluctantly for 2 reasons, reason one is I'm not a mastering engineer and I shouldn't be mastering my own mixes, sort of defeats the idea of a second set of ears. Reason 2, and the biggest reason is that without fail the customer comes back and tells me the CD is too quiet. I don't own a limiter, and I never will. I use compression sparingly. My normal method for getting the loudest CD I feel comfortable putting out is to compress the mix about 4:1. Then I normalize the track in Logic, seek the stray loud peaks and reduce them a couple db, I do this several times, and re-normalize. It's never enough for the client. It's a shame to hear the decline in fidelity that's happened over the last couple of decades.
I really appreciate you taking the time to illustrate this problem, hats off to you and keep fighting the battle. I haven't had time to read the other comments yet, but I will. Best of luck to you.
Mike Oswald
www.sdpstudio.com

By Mike Oswald on Saturday, February 09, 2008 1:51 AM

Re: Over the Limit

I've re-visited this article a few times; it remains pretty fucking kickass. Sorry, younger generation.

Great writing, and you know your shit, man.

Thanks for putting your words out there.

By Mogul of Crabs! on Tuesday, March 11, 2008 9:53 PM

Re: Over the Limit - Required

Thanks Rip.
Your article will be required reading for my Audio Engineering II class (among others). While there isn't anything that hasn't been said before, you do a great job of getting the facts and opinions separated, and (of course) you use a great band for subject matter.
I use the title track from Counterparts as an example of 'the limit' of loudness and with its level being about K12 (I recommend K14). Regardless, for the most part, enough room for the majority of transients and loud enough to 'compete' with some CDs out there. One will note that in this track it sits at about K12-10 but 'drops' to K13 and 14 and pushes all the way to K8 depending on the section of the song. In general, the level 'rides' a bit and gives the listener ups and downs. K12 is also about the limit where the drums start to get pulled backwards depending on the RMS to Peak ratio of the drums themselves and frequency range of the voice (the main lead instrument).
Mike's method of" normalize . reduce peaks at the track level, normalize . reduce peaks at the track level is a good one. However a statement of 'compress the mix at 4:1' doesn't mean anything since the ratio and threshold go hand in hand and without that knowledge you can't tell. A ratio of infinity:1 would be transparent if the levels only crossed the threshold sparingly. Average Gain Reduction (the meter) will tell you a lot more. Thus a ratio of 1.1:1 with a very low threshold will produce a higher GR than something with a seemingly high ratio. If you think about it, 4:1 is pretty freaking severe if you have a transient of 20dB (over) you only get 5dB back. AGR will give you a number that means something.

By Paul Bissell on Wednesday, March 19, 2008 7:51 AM

Re: Over the Limit

In my opinion, Vapor Trails party sounds like a big untolerable mess in most places because there wasn't a proper producer there with a second opinion to tell the guys not to triple track bass lines in the same register. The parts of the CD with a single bass line sound great; otherwise all that power in the bass spectrum is just overwhelming.

By Fred on Sunday, March 23, 2008 10:57 AM

website page counter