Music Into Noise: The Destructive Use Of Dynamic Range Compression part 2

color saturation rules fuiud

Anyway, there's nothing wrong with compression, it's really nice to thicken up a mix but it does get abused to the point where it's just limiting sonic potential imo, that's the only thing to lament really.

― historyyy (prettylikealaindelon), Sunday, October 30, 2011 4:38 PM (14 minutes ago) Bookmark Suggest Ban Permalink

you're confusing things. there is something wrong with 'dynamic range compression' which is what we're talking about

cropping the frame so it fits on tv

― zvookster, Sunday, October 30, 2011 3:39 PM (1 hour ago) Bookmark Suggest Ban Permalink

That's a good one.

― Tim F, Sunday, October 30, 2011 4:25 PM (28 minutes ago) Bookmark Suggest Ban Permalink

this is the most OTM comparison -- pan & scan cinema is the movie equivalent, if there is one.

i mean i suppose the real equivalent would be streaming netflix on a bad connection

I don't know dude, Godard rocked heavy saturation hard in parts of In Praise of Love. It looked pretty good to me.

Sure, it CAN look good in doses (or as you say, "in parts"). But would you want everything you watch to always have the levels cranked ALL of the time? And even if you thought you DID, would your eyes and brain appreciate it?

color saturation is definitely something you can want all of the time

http://languages.oberlin.edu/courses/2011/spring/cine299/mmeyer/files/2011/02/in-the-mood-for-love2.jpg

xp Actually I think pan and scan is partly back to data compression, i.e. losing parts of the data that aren't seen as important to get the gist across. So I'm not sure it's really the analogue of DRC here.

What I like about the "cropping" analogy is that both are a pragmatic solution to the demands of modern modes of consumption. Obv. there is a basic advantage to cropping the frame if you are watching a film on TV; just as heavy DRC makes sense for commuter and radio listening.

It's less cropping and more like stretching a 4:3 image to fit a widescreen tv. Many, many people don't care at all but it drives me batty.

Yeah, that's probably closer to the nub of it - all data is still there but it's distorted somehow. I'm not sure there's a perfect analogue here because with visual media it's pretty clear there's something up, and with audio it's far less obvious.

I don't agree. Altering volumes over the course of a mix is not a distortion like stretching an image.

And I'm not clear on deej's distinction. "Compression" is "dynamic range compression" as far as I know.

I don't agree. Altering volumes over the course of a mix is not a distortion like stretching an image.

― timellison, Sunday, October 30, 2011 5:22 PM (10 minutes ago) Bookmark Suggest Ban Permalink

dynamic range compression is not the same as 'altering volumes'...

its 'altering them so that they distort'

No, there was a problem with clipping in some cases of mastering with extreme compression, but the use of compression does not necessarily entail an introduction of distortion.

fwiw my use of "distorted" in previous post is referring to distorting the original dynamics, not distortion as in peaking.

Right, but when I'm mixing my own music and I alter the volume level at a particular point on a given instrument, I don't feel like I'm "distorting the original dynamics." In fact, I only care about the original dynamics to the extent that they sound good to me.

"Compression" is "dynamic range compression" as far as I know.

yes, not sure what deej is getting at here.

its 'altering them so that they distort'

I guess if you were clipping then you'd be distorting, but that's not compression.

or rather i should say not the aim of compression.

I get what you're saying Tim but there's a difference between making those decisions during mixing, where you're consciously making the choice to make certain things quieter than others (or louder than others) at specific points in time, and just taking a blanket algorithm and making even the "quiet" parts sound as loud as the loud ones, no?

Saying "compression is fine" or whatever you said in your original post is missing the point;we are talking about boosting levels so they become compressed in the mastering stage - not as a tool of production, but a way of mastering something LOUDLY. That isnt fine at all, it's pretty uniformly an awful choice

Dynamic range compression does cause distortion...thats what makes it bad. It flattens relative dynamics, distorting the sound of a track. It's not the same as using compression as a tool in production

there's a difference between making those decisions during mixing, where you're consciously making the choice to make certain things quieter than others (or louder than others) at specific points in time, and just taking a blanket algorithm and making even the "quiet" parts sound as loud as the loud ones, no?

Absolutely, but I'm a little conflicted on it. On the one hand, I think flatness in dynamic range can be a positive in pop records - that was why I brought up Spector and Motown the other day. I tend to think I'd like to see something like that arrived at organically rather than "applying an algorithm," but I guess I'm open to the idea that applying that algorithm could make it sound better than what I came up with when I mixed it.

xp, I think you're using the term distortion quite liberally, it's not actual distortion. Compression makes tracks as loud as possible up until clipping starts to occur, technically I don't think it qualifies as distortion. Compression is a tool used by mastering engineers and musicians, in both cases it flattens dynamics, it's the same thing. I think we agree that the issue is the trend that governs how mastering engineers employ it.

clipping is digital distortion

I'm not convinced reviving this thread is the best idea I've ever had, but this Monolake example made me chuckle:

http://soundcloud.com/monolake/mastering-a-step-by-step-guide

Ha! That's evil.

worthy bump

I lol'd, but not too loud of course

Musley 2 days ago
WHERE IS THE DROP?

awesome

love it

Monolake is an all-time cool dude.

hope everyone is as psyched as i am about dynamic range day next month!

http://dynamicrangeday.co.uk/

wow that record that won last year is really bad

So I'm being interviewed over the phone tonight by a Brazillian newspaper guy for a piece he's writing about loudness in pop albums. Which is strange.

http://i.imgur.com/Rc4Ag.jpg

Ha!

Current Field Music album sleeve contains the message "In order to preserve sonic fidelity, this record has been mastered using significantly less compression and limiting than most contemporary records. For maximum listening pleasure, please turn your stereo system UP!"

apologies if this has already been posted

a measured discussion with the always subtle bob ludwig of gateway mastering

http://thetrichordist.wordpress.com/2012/06/28/recording-tips-for-the-loudness-wars-an-interview-with-bob-ludwig-of-gateway-mastering/

He's pretty far off the mark in saying kicks and bass mixed up the middle came about as a response to earbuds. Dance music, buddy. Like from way back when.

maybe those are his own personal reasons

I remember reading the dude from Liquid Liquid saying you could tell "White Lines" was mixed for radio whereas "Cavern" was mixed for clubs because of the way the bass was mixed on each record, midrange-y on the former, and a much deeper tone on the latter.

Futher monolake goodness here - http://roberthenke.com/interviews/mastering.html - it's a discussion between him and Rashad Becker of Dubplates and Mastering about mastering and covers some issues related to dynamic range compression. As far as I can tell, this discussion seems to be assuming classic status as a comprehensible and informative guide to what mastering is and why it's important.

Shellac's Bob Weston gives his 2p here - http://www.chicagomasteringservice.com/loudness.html

http://www.nature.com/srep/2012/120726/srep00521/full/srep00521.html

In the excellent interview with Susan Rogers, mid-80s Prince engineer, that's been making the rounds today, a great Q/A bit:

DRS: Speaking of recording techniques, when I listen to a lot of music these days, it seems really loud. It’s not just a matter of ‘things have progressed in technology,’ but it just seems like the music is just louder for the sake of being loud. Maybe it’s just another sign I’m getting older (laughing). Am I off base with that? As an engineer who’s been around for years, what do you think?

SR: I know what you’re saying. The technique these days involves hyper-compression where in mastering, and sometimes even before, you squash out all the dynamics. You level the dynamics such that there’s no change in loudness going from the verse to the chorus and the climaxes of the song don’t get any louder than the quiet parts of the song. The trend began in the ‘90s…it originated from radio broadcasters who wanted program levels to be uniformly loud. They didn’t want any quiet moments that might allow a listener to switch to a new station, so record makers started competing in the same way by flattening out the dynamics so that your record would be louder than the next guy’s…and it sounds great when you put your record on and it just comes in hotter than the next person’s. We know, at least here in the Western world, consumers prefer whichever audio source is louder. It can be a fraction of a DB hotter and the consumer will say “yeah, that one sounds better.” But what has happened, by reducing these dynamics we’re actually changing the emotional impact, (I’m arguing this anyway) of musical material because dynamics are what gives a song tension and release…it gives it a payoff. To take away the dynamics, you can listen longer because there’s nothing changing so you can listen for a longer period of time but you’ll probably be less emotionally engaged than you would have been otherwise. Dynamics contribute to emotion, but, that said, we are now writing and producing music such that you don’t need a lot of dynamics. It’s changing the way composers and producers are working. How we think of music nowadays we think of it as being kind of uni-dynamic.

Preach

There is no compression problem in modern music. And complaints to the contrary seem awfully close to the incident when Paul McCartney's dad suggested that Paul change the lyrics the line "yeah, yeah, yeah" to "yes, yes, yes." Because that's the way it SHOULD be.

This is one of their issues where I respect the people making the argument but I feel they're ultimately misguided (not unlike Wayne Coyne complaining about copyright infringers). The idea that "too much" dynamic range compression objectively makes music sound "bad" is kind of creed or article of faith that doesn't make sense on closer inspection. What amount of compression is "too much?" How much dynamic range is "enough?" Even the professional audio engineers can't agree on this or every good record from 1980 would sound the same as every other good record made that year.

Posting a few images of waveforms doesn't make your argument any more consistent or logical. If you posted a waveform for a orchestral classical piece, it would make most popular pieces seem sorely lacking in dynamics - even jazz would "suffer" in comparison. (King Crimson is the only rock band I can think of with dynamics even approaching the classical genre, at least in the early days).

I've met people invested in classical music who do indeed deride rock albums for their lack of dynamics. However, I think this has more to do with their personal taste than any inherent problem in rock mixing or engineering. Super-compressed recordings have a unique sound and I guess you either like it or you don't. What bothers me is when people start making claims that too much compression objectively reduces the level of listener enjoyment - "it inevitably tires out your ears, causing you to unconsciously tune out and miss any subtleties in the music." I know not everyone says stuff like that but it's pretty common in this topic.

Old records from the 1920s and earlier had virtually no dynamic range, right? And yet people still enjoyed them well enough. Also, a lot of musicians and music fans have bad hearing, which can greatly reduce heard dynamic range. Are they objectively enjoying the music less than those of us with fine hearing?

Do you really not think that bad hearing could reduce your enjoyment of music?