Compression

dHarry

Well-known member
It's not as simple as [compression=bad] - put simply compression is equal to turning the volume up for the naturally quiet bits so that more of the sound is close to the max volume available, and there is less difference between the loud and quiet parts.

It's been around as a special effect since at least the 60's - Ringo Starr's crash cymbals spring to mind as obvious examples - they were always compressed so that the crash tail-off is louder and longer than "natural". See also Led Zep's When The Levee Breaks, where the hi-hats between beats become as loud as the kick/snare, creating that huge sound. I think it's a crucial factor in the original Amen break also. Also on Daft Punk's One More Time where the brass/synth sound rises up in between kick drum beats due to compression - the speed of the compression kicking in adds crucially to the pumping/breathing dynamism of the track (also used on the synths on Madonna's new single).

The problem is when the final master track is compressed to the max, so as to sound as LOUD AS POSSIBLE ALL THE TIME and dynamic range is lost - the Killers I think do this a lot, sounding extra loud on the radio, but ultimately inducing listening fatigue.
 

michael

Bring out the vacuum
Weird synchronicity, I was listening to something hideously overcompressed today and was thinking about mentioning it on here tonight! So, yeah, I definitely think this article is on to a good thing.

In the recent Wire interview with Scott Walker he sums everything up, IMO:

Scott Walker said:
We're a bit purist in the sense that we don't use a lot of compression on our records, once again, it seems to cramp up the space and everything sounds very flat. There's so much compression on everything today, just to make it loud. Everybody wants to be louder than everybody else. Then when they get it to the radio station, they add more compression on it, the disc jockeys, cos they want to be loud and the station wants to be loud. So you have this flat, cramped, digi-noise. So, when you listen to our records the best thing to do is really crank them up loud, cos then you'll hear everything. You'll hear all the space and it'll be comfortable.
I guess it's purism, but when you're talking about diminishing the ability of a recording to have any kind of dynamics-related impact then I think it's pretty easily justifiable. It's not like saying you must use a certain kind of recording set up or something, just saying to take it easy wrt a particular technique.

Really extreme compression has been used a lot to interesting effect, what's called "ducking" and things like that, but that's usually applied to particular elements in a track, not a whole recording.

I thought this explanation of a whole bunch of sonograms and the visual impact of compression was pretty interesting:

http://www.airwindows.com/analysis/Albums/index.html
Or go straight to his Bad Current Examples

I certainly can't agree with this guys summation of the records he talks about, but whatever.
 

IdleRich

IdleRich
I agree that he takes a long time to say the same thing again and again in the article although the points do build up to make his argument more persuasive.
I also know that compression isn't automatically bad - I love it when it's used as a tool to make the song sound better, the classic example being Joe Meek (especially with the Honeycombs) but if it is being used to bland out the peaks and troughs of everything as a default sound that's got to be bad. If what he says is true I find it rather depressing, the way it seems to have happened due to the way people listen to music also seems to link with what someone said in the grime thread about production not needing to be that good because tracks are mainly going to be listened to through a mobile phone.
 

IdleRich

IdleRich
Thank you Edward.
What I don't get is why don't artists act to stop their songs being "clipped" like this? Is it because a) They don't notice b) They don't care c) They have no control? None of those things bode well for the music of the future.
 

hint

party record with a siren
IdleRich said:
What I don't get is why don't artists act to stop their songs being "clipped" like this? Is it because a) They don't notice b) They don't care c) They have no control? None of those things bode well for the music of the future.

It's been drilled into so many artists that louder = better. They notice it when it isn't there - "Dude! Our album needs to be as loud as Metallica's!".

Most mastering engineers are opposed to the "Loudness Race", but continue to compress as much as they do because that's what you're supposed to do in that line of work - give the client what they want.
 

Logos

Ghosts of my life
There is a problem with brickwall limiting in pop music these days, part of it I'm sure is the sonic arms race about making your record sound as loud as the last tune played on the radio.

But compression related techniques in house music specifically, have developed as a positive sonic device too - the whole pumping daft punk sound wouldn't have existed without sidechaining (a technique that 'ducks' the kick drum when the kick and bass occur at the same time) and the generally squashed mixdown, as well as those filters of course.

My point is there are always creative ways of using techniques that on paper are destructive, or antithetical to coventional audiophile virtues.
 
Last edited:

hint

party record with a siren
Freakaholic said:
this, in my opinion, is the main reason for the downfall of punk and metal: overproduction, overcompression.

Indeed.

I wouldn't call it "the downfall"... but I certainly perfer it when any type of rock music keeps those dynamics intact.
 

Grievous Angel

Beast of Burden
Louder tracks when heard on the radio sell more, hence the loudness arms race.

Not all compression and limiting is bad; used well they are important weapons in an artist's armoury.
 

bassnation

the abyss
2stepfan said:
Louder tracks when heard on the radio sell more, hence the loudness arms race.

Not all compression and limiting is bad; used well they are important weapons in an artist's armoury.

particularly where subbass is concerned - compression is very important for that, if you want loud, punchy sub that doesn't overwhelm all the other elements. of course, this is very different to compressing the entire track.
 
Last edited:

Slothrop

Tight but Polite
bassnation said:
particularly where subbass is concerned - compression is very important for that, if you want loud, punchy sub that doesn't overwhelm all the other elements.
And on drums, and pretty much everything else. It's the difference between compressing a lot of the elements of the track to make them sound tighter, and compressing the fuck out of the whole track to make it jump out of the radio more at the expense of any dynamics.

I don't think anyone's down on compression as a whole...
 
there's a study being done in the States about

the effect of shredding (extreme limiting, more or less clipping, the look of a waveform when processed this way) on listeners turning away from radio I heard...
 

SIZZLE

gasoline for haters
A pretty simple reason for why this is perpetuated at all stages of the process is that to the un-trained, and sometimes trained ear louder simply sounds better. It's sounds 'more detailed' 'punchy' etc. So very often a mastering engineer's job is simply to gain as much level as possible by any means necessary, whether that's squashing peaks or filtering out inharmonic frequency bands. A lot of people have mentioned radio but this also EXTREMELY perhaps more important when mastering and cutting club records. If your record, when mixed in by a DJ at the same gain as the previous record causes a drop in overall level it will sound weaker, especially in genres that privilege stuff like powerful bass. Everybody wants their tune to sound better than the tune that came before it, hence the loudness race.
 

hint

party record with a siren
DJs (radio and club) need to be reminded where the gain knob is.

A dynamic recording turned up wipes the floor with any über-squashed "loud" track.
 
Last edited:

Lichen

Well-known member
Ad Breaks

Is compression the reason why TV ads are always 25% louder than the programs they interrupt?

Do ad. producers compete for volume?


I know that TV Channels promise they don't up the volumes.
 

hint

party record with a siren
Yeah - it's compression that's at work there.

"Loudness" is all about average volume. By reducing the dynamic range through compression, you make it possible to increase the average volume.

The TV companies are correct - the highest peaks during adverts aren't any higher than those during normal programmes, if you look at it in terms of dB.

The difference is that, during the adverts, the audio is nearly always peaking... so the brain interprets this as being louder.
 

DigitalDjigit

Honky Tonk Woman
Thank god for compression. It rescued us from rickety drum sounds in techno. Whatever you may say about hard loopy techno since '98 or so, the impact of it is so physical, it just grabs you by the balls.
 
Top