|
Loud, Louder, Loudest: The Tradeoff of MP3s
|
View this Thread in Original format
| stevieboy32808 |
I just bumped into a great article in the Wall Street Journal regarding the the intricate differences in sound of a compressed music file versus an uncompressed studio recording. It's very interesting and sheds some light on how record labels master cds for the ipod generation and so forth.
| quote: | Are Technology Limits In MP3s and iPods
Ruining Pop Music?
September 12, 2007
by Lee Gomes
Wall Street Journal
If it seems like you are listening to music more but enjoying it less, some people in the recording industry say they know why. They blame that iPod that you can't live without, along with all the compressed MP3 music files you've loaded on it.
Those who work behind-the-mic in the music industry -- producers, engineers, mixers and the like -- say they increasingly assume their recordings will be heard as MP3s on an iPod music player. That combination is thus becoming the "reference platform" used as a test of how a track should sound. (Movie makers make much the same complaint when they see their filmed images in low-quality digital form.)
But because both compressed music and the iPod's relatively low-quality earbuds have many limitations, music producers fret that they are engineering music to a technical lowest common denominator. The result, many say, is music that is loud but harsh and flat, and thus not enjoyable for long periods of time.
"Right now, when you are done recording a track, the first thing the band does is to load it onto an iPod and give it a listen," said Alan Douches, who has worked with Fleetwood Mac and others. "Years ago, we might have checked the sound of a track on a Walkman, but no one believed that was the best it could sound. Today, young artists think MP3s are a high-quality medium and the iPod is state-of-the-art sound."
It isn't. Producers and engineers say there are many ways they might change a track to accommodate an iPod MP3. Sometimes, the changes are for the worse.
For example, says veteran Los Angeles studio owner Skip Saylor, high frequencies that might seem splendid on a CD might not sound as good as an MP3 file and so will get taken out of the mix. "The result might make you happy on an MP3, but it wouldn't make you happy on a CD," he says. "Am I glad I am doing this? No. But it's the real world and so you make adjustments."
This shift to compressed music heard via an iPod is occurring at the same time as another music trend that bothers audiophiles: Music today is released at higher volume levels than ever before, on the assumption that louder music sells better. The process of boosting volume, though, tends to eliminate a track's distinct highs and lows.
As a result, contemporary pop music has a characteristic sound, says veteran L.A. engineer Jack Joseph Puig, whose credits include the Rolling Stones and Eric Clapton. "Ten years ago, music was warmer; it was rich and thick, with more tones and more 'real power.' But newer records are more brittle and bright. They have what I call 'implied power.' It's all done with delays and reverbs and compression to fool your brain."
All these engineers tend to be audiophiles, the sort who would fuss over a track to make it perfect. But they're beginning to wonder if they should bother.
"I care about quality, even though the kid on the street might like what he hears on MySpace, which is even worse than an MP3," said Stuart Brawley, an L.A. engineer who has recorded Cher and Michael Jackson. "We try to make the best quality sound we can, but we increasingly have to be realistic about how much time we can spend doing it."
Howard Benson, who has done work for Santana and Chris Daughtry, says members of a studio recording crew will sometimes complain after a session, "I just spent all this time getting the greatest guitar and drums solo, and it ends up as an MP3."
Even those who complain about MP3s say they own and enjoy iPods, and appreciate how they have made music so widely available. They just wish, they say, the device wasn't setting the technical standard for how music gets made.
Of course, not all music producers agree that MP3s and iPods are affecting music in quite so bad a way. Larry Klein, noted for his work with Joni Mitchell, said, "If something sounds really good on an average pair of speakers, it will sound great on earbuds. I can't imagine mixing a record so that it sounds better on earbuds."
And Clif Magness, who has recorded with Kelly Clarkson and Clay Aiken, says music recorded by young artists in living rooms via MP3s, while technically crude, can sometimes have an urgency and immediacy that might be missing from slick studio projects.
When CDs were first introduced, they were regarded as cold and flat, compared with vinyl. But their sound improved as engineers learned the medium, a process many hope will happen again with MP3s and portable music players.
Michael Bradford, who has produced Kid Rock, notes that as storage and bandwidth capabilities grow, music won't need to be as compressed. Even now, some audio buffs, such as Stereophile magazine columnist Michael Fremer, insist on a best-of-both-worlds approach to digital music. He uses $500 earbuds with his iPod to listen to digital, but uncompressed, music he captures from vinyl LPs.
Still, engineers experience some nostalgia about earlier technologies. Says Mr. Saylor, "What we've lost with this new era of massive compression and low fidelity are the records that sounds so good that you get lost in them. "Dark Side of the Moon" -- records like that just aren't being made today."
Source: http://online.wsj.com/public/articl...6892024096.html |
This is what it boils down to: http://www.youtube.com/watch?v=3Gmex_4hreQ |
|
|
| MrJiveBoJingles |
| Giving the people what they want. :clown: |
|
|
| MrJiveBoJingles |
| It's the fast food attitude toward music. As long as it makes their mouths water and gets their money in our pockets, right? |
|
|
| Sykonee |
Been bitching about this for a while now, but then I tend to make use of headphones that will bring out MP3 limitations. I find a 192kbs format fairly good though.
But yeah, the whole "crank that volume so it'll sound good on crap iPod buds" bit makes my blood boil. You're just getting assaulted by a wall of sound that lacks dynamics, which is only good if the music is intended to be listened as such (like noizecore, heh).
Folks should really take a listen to Brian Wilson's Smile to hear how lovely music can sound even today using older recording techniques. |
|
|
| Clovis |
| quote: | | Today, young artists think MP3s are a high-quality medium and the iPod is state-of-the-art sound. |
Entirely their fault. |
|
|
| a98 |
extreme compressing and limiting has been a problem for few years already. there seems to be a competition who has the loudest tracks. very sad indeed, but what can you do. i do hope that one day they reach the max (when all the masters just look like a rectangle) and they decide that dynamics are more important than the volume.
i've personally tried to keep my collections at decent levels. editing all too much limited tracks to lower volumes. so that they are as "loud" as other tracks in my collection, but ofcourse sound tier (messy mixing etc) since someone killed the dynamics. |
|
|
| Existo22 |
This is relevant to rock jazz and acoustic genres.
These genres want to capture performances in live rooms and making the recording too loud makes the track sound artificial.
But electronic music has always been artificial and mechanical and has always been loud.
Always.
On EDM the sounds and samples are already very very compressed. This is the nature of the dance genre.
Step sequencers... akai samplers 808s 909s SP1200s.
It is also going on vinyl so it is important to get the track sounding as loud as you can to cover up the hiss. It will either be played on a radio which is as loud as sound gets (The louder the sound the further they reach)... or in clubs.
In either case a brickwall limiter is on the output pulled all the way down...
What makes a track sound good is the quality of the gear that is pumping volume into the music. While my waves L2 is nice it certaintly won't compare to the 20.000 limiters they have at the sony mastering studios.
Gear for the radio and broadcast market costs a fortune because they know that they will push the sound into clipping. It is important that that happens in a way that is musical and when you are shooting the superball for ex. to 50.000.000 people there is no room for compomise you need the best regardless of the cost.
This is why the engineers mixing the britney spears cd can push the out of it and still make it sound warmer than your production.
It is good that you read up and learn about stuff like ''the loudness war'' but know when that info does not apply to your genre coz arguing that ''our trance track does not sound as natural as the room and instruments we miked up to capture the trance performers'' or that ''we are losing the warm fidelity and dynamics of trance'' makes you look silly. |
|
|
| SuspicionVandit |
if they are going to talk about ipods, they might as well discuss AAC :whip: which is
1. the codec used for music from iTunes
1a. superior to MP3. |
|
|
| Derivative |
| quote: | Originally posted by a98
extreme compressing and limiting has been a problem for few years already. there seems to be a competition who has the loudest tracks. very sad indeed, but what can you do. i do hope that one day they reach the max (when all the masters just look like a rectangle) and they decide that dynamics are more important than the volume.
i've personally tried to keep my collections at decent levels. editing all too much limited tracks to lower volumes. so that they are as "loud" as other tracks in my collection, but ofcourse sound tier (messy mixing etc) since someone killed the dynamics. |
No it hasn't.
You haven't heard what overcompression is until you have got a compressor, run it before the output stage of your amp with a threshold of -60dB (or thereabouts) and a compression ratio of 10:1 (or thereabouts). Then normalise the output to 0dB. Be sure to turn down your amp output level ALOT because it will be loud.
That is overcompression. If you think you have heard anything like that in professionally released music you have got to be out of your mind.
Compressors are tools that control gain. A paintbrush is a tool that controls the application of paint. Like a palette knife.
How you use them is a matter of preference and appropriateness with respect to what has to be done.
You can no more say modern music has too much compression as you can say that William Faulkner 'doesn't use enough punctuation'. Or Mark Rothko 'doesn't use enough colours.'
So when I hear this bull repeated for the nth time I am reminded of that scene in Amadeus where Mozart is told 'your work is good but it has too many notes'. To which Mozart replies perplexedly 'but sir, the peice neither has too many nor too few notes. It has exactly the number of notes required. No more, no less.'
If you are talking about high level outputs in clubs then that is another issue. You use limiters at the end of the signal chain to prevent stray clips from destroying the soundsystem. The people listening to the music at 116dB wear properly fitted earplugs - as any clubber should because any sound at 116dB is destroying your middle and inner ear progressively in a matter of minutes. You would be a fool not to do this.
But as soon as you have people in 'the industry' telling you what is and is not creatively acceptible you have the end of music as art. Its even more hilarious when engineers say stupid like 'music used to be warmer and fuller with more tone'. All I read is setence full of onomatopaeic words which don't describe sound adequately. |
|
|
| a98 |
| quote: | Originally posted by Derivative
stuff.. |
you can say that a chainsaw is very noisy though it's no where in the same level as a jet plane or a bomb for instance.
for some people it doesn't have to be that much limited it to become too limited. for me if the RMS values are more than -12dB, the track is overlimited. and even -12db is kinda too much, -13dB is enough.
one track i recently bought had the rms in -9.5dB, you can imaging how loud it is.
the spectrum looks like this:
 |
|
|
| Derivative |
| quote: | Originally posted by a98
you can say that a chainsaw is very noisy though it's no where in the same level as a jet plane or a bomb for instance.
for some people it doesn't have to be that much limited it to become too limited. for me if the RMS values are more than -12dB, the track is overlimited. and even -12db is kinda too much, -13dB is enough.
one track i recently bought had the rms in -9.5dB, you can imaging how loud it is.
the spectrum looks like this:
|
This is where you up. That image you posted is from Sony Soundforge and the waveform is zoomed all the way out. Bear in mind that you can zoom in to view individual samples. In a 44.1khz recording that is 44,100 samples per second. You will note that if you zoom in only a small degree you will begin to make out individual transients much better and even electronic dance music has massive dynamic range. I'll post some example pictures in a while to show you exactly what I mean.
About RMS and the idea that 'RMS peak level is rising'. What do you expect for a hard rock recording or a dance anthem? Oasis's 'Definitely Maybe' is a loud, brash rock recording and it is recorded in a loud and brash way.
Do you expect Oasis to make a record that is normalised to -3dB with no compression? Back in the day Noel Gallagher used to just swipe all the volume and EQ dials on his amp to 10. It is telling because all of this is documented in the film 'There we were now here we are' and they also note that they went through about 10 engineers before they found one who was actually capable of producing something which captured what Oasis sounded like live. And that guy brickwalled every track.
You also realise that there are no standard measurements for RMS peak level calculation seeing as the window in which peak level detection occurs and is averaged varies massively? In soundforge you can vary the size of this window from 5ms to 1000ms. Thats a huge difference when calculating average peak level and the results will vary massively depending on the size of said window. Some meters and dynamic processors have RMS peak level detection that goes up to 3000ms.
Also you have a fundamental misunderstanding of what sound pressure level is.
You don't need to drive an ear bud headphone with anywhere near as much power as you do a full PA system but many earbud headphones are capable of generating exactly the same sound pressure level in the ear canal causing exactly the same amount of damage to your inner ear.
A chainsaw, a plane, an earbud headphone. It doesn't matter what it is. All it needs is close enough proximity to the ear to generate the same sound pressure level in the ear canal and it causes the same damage. It can also be safely listened to for the same amount of time before it starts permanently damaging your ear. For an SPL of 116dB in the ear canal you are talking a couple of minutes at most before it starts ing your middle and inner ear. How that pressure level is generated is meaningless.
So if you expose your ear to this kind of SPL on even an occassional basis you should wear fitted earplugs. |
|
|
| a98 |
| quote: | Originally posted by Derivative
This is where you up. That image you posted is from Sony Soundforge and the waveform is zoomed all the way out. |
only way to see the difference in the spectrum is when it's zoomed to full size. one small part of the song doesn't tell anything about the overall dynamics. and even then the visual doesn't really tell anything, i just wanted to show how it looked.
recording something loud has nothing to do with the overall dynamics of the track. mixing and mastering process is what counts. and rock music is rock music, they can do what ever they want i don't care.
and a track being a dance anthem doesn't mean it has to be limited too much. it will probably sound a lot better if it has more dynamics when it's played loud in a club.
and ofcourse mastering/overlimiting can be a form of art too, unfortunetly in dance music 99% of the case the artist doesn't get to choose how the track is mastered. so if a label decides to compress it flat, it isn't nessasery what the artist wanted from the track.
my point was just that overlimiting ruins the songs in my opinnion. if the mixing hasn't been done perfectly, it will make the song sound really messy. ofcourse it's something the artist/label decided to do, but that doesn't mean i have to put up with it and listen to that.
too much limited/compressed track has to be really good for me to still wanna enjoy it.
but yeah whatever, i don't wanna argue with you anymore, we're clearly on the wrong page. and you seem to be an expert on mastering though you've probably never studied or done it professionally. |
|
|
|
|
