R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1]   Go Down

Author Topic: Some thoughts on Compression and Loudness by Richard Kaplan of Indigo Ranch  (Read 2114 times)

studiorathq

  • Newbie
  • *
  • Offline Offline
  • Posts: 1
  • Real Full Name: Richard Kaplan

"It's one louder" Nigel Tufnel of Spinal Tap

This article is reproduced from my website www.studiorathq.com  Waveform images are available on this topic in the studio rat forum.
 
I recently came across the above cartoon posted by a friend on Facebook and it humorously makes a good point.  How do you make your recordings louder than others?  But also how much is too much?   The main process to get your recordings 'louder,' other than recording the incoming instruments as close to "peaking" the input meters is to use compression and limiting.  It also matters how it is mastered, but let's focus on compression in the recording and mixing process.  Compression has become an essential component in today's recordings.  I've been in the music business for over 30 years and I've watched this fight for louder recordings continue again and again.  I've worked in all genres of music, but I see it happening mostly in Metal and Rap. 
 
In the early days of recording there were really only a couple of effects that the engineer could use.  One was tone control - which later became EQ - and by tone control I mean a turn pot, like on an old car radio, that boosted the bass when turned to the left and boosted the treble when turned to the right.  The other effect was volume control which was done in a number of ways - starting in the beginning with the engineer placing the ONE microphone at a distance from the various instruments, in order to get a balance, to then go to the ONE mono track on which everything would be recorded. Then came multi track recording and the engineer could move faders up or down!!

Finally came the compressor and limiter.  When you think about it, a compressor or limiter is really just a box full of electronic parts with a job similar to that of an engineer.  Visualize this imaginary engineer manipulating a fader to avoid letting a signal that was too loud to go to tape, that is the primary function of a compressor/limiter.
 
A limiter - just as the name suggests is a device that absolutely limits the signal from going above an established level (voltage or VU). When the signal reaches a certain point the 'hand in the box' limits the level so that it NEVER goes above that agreed upon level.
The limiter was really born from the need to smash anything above a certain level, with the fastest hand possible (the imaginary engineer in this electronic box) to avoid over-modulating a radio signal.  There were (and still are) severe monetary penalties if an audio signal goes over a station’s allowed wattage when transmitting.  This process of limiting also became handy for recording when the engineer needed to keep the signal from distorting as it is recorded to tape.
 
The compressor is similar but it gave the engineer more control.  Often times what is called a limiter is, in reality, a compressor and vice versa. The compressor (and some limiters) allowed the engineer to select a level or VU above which the device started to take control of the imaginary fader in the box. This is known as the “threshold” control.  It also gave the engineer a ratio at which the imaginary hand in the box brought down the fader. This is the compression “ratio.” Often there are other controls that dictate how quickly the imaginary hand in the box responds to sudden increases in level; this became known as “attack.”  Another control variable that dictates how quickly the imaginary hand it the box brings back the level is known as “release.”
 
These were originally meant to keep the engineer out of trouble with over-modulating and distortion.  Later it was found that these devices could be used as an effect! . . Effect?!  Yes, especially if the trigger to the threshold was something that pulsed up and down with the beat in a certain section of a song (verse, chorus, bridge etc.) or throughout the entire song.  That would make the compressed item appear to "suck & blow" rhythmically with the music. It could be used to rhythmically pulse one or many instruments or even the entire song. This was really the first of many effects put in the arsenal of the recording engineer.
 
But then what happened?
 
Radio stations realized that by using an almost unnoticeable amount of limiting they could sound louder than competing stations using the same legally allowed wattage!  Mastering engineers found that a little bit of limiting could make their vinyl records sound louder than other similar-sounding records; and even multiply that apparent loudness when put on a radio station with that aforementioned limiting.  Soon it was realized that the music signal could be split into bandwidths (filtered by frequency) that could then be individually limited, creating an effect of added bass or “presence” to their programs.
 
The RACE to LOUDNESS was on!
 
This has come to the inevitable conclusion that we can hear, and are now commenting on, in many modern recordings.  You will notice that it is hard to unravel exactly when in the recording process this desire for greater loudness took place.  It could be in the original recording of the song.  It could be in the mixdown of the song.  It could be in the mastering.  Now with most recordings being completed in the digital domain it is even easier to get locked into wanting to win this loudness contest.
 
There are so many digital plug-ins that approximate nearly every limiter and compressor ever made, in addition to a whole host of never before heard of devices, some of them even offering zillions of micro controls to this loudness game.
 
We are posting a few waveforms that show clearly what can happen to a song if this goes to the extreme. Most of these examples seem to indicate that the recordings received the worst blow in the mastering process.  The problem is that the life and natural pulse has been removed from the music; making it sound sterile without the essential dynamics that give music and songs their life.  Most listeners agree that, before compression, mixes actually sounded more pleasing; although ultimately not as loud.
 
Seeing and hearing the difference!
 
Where can you hear this process in action?  One example is Metallica's last album: Death Magnetic.   This record was a major release, recorded on an unlimited budget, with the greatest equipment in the industry.  Death Magnetic was produced by Rick Rubin. . . a man I've worked with at Indigo Ranch.
 
You may or may not like this band or album, but upon the release of Death Magnetic the Internet was littered with fans excited about Metallica's return to classic form, but the net also had an equal number of complaints about the quality of the sound and production.  Rubin and Metallica's intention was to push the boundaries of recording; to make a record that was at the forefront of the loudness curve.  They probably succeeded, but they also paid the price for pushing the curve.
 
Sean Michaels of The Guardian explains that on Death Magnetic: "the sound issues are a result of the 'loudness war' – an ongoing industry effort to make recordings as loud as possible."  The album's "sound issues" are as Wikipedia points out the result of: "overly compressed dynamic range with a process called peak limiting leading to audible distortion."

The cd version of a Death Magnetic track allows no room for dynamics and this mix is ‘brickwalled’ by limiting.  After listening to Death Magnetic many times, it left me wondering:  Do the instruments mix well among themselves?  Would I want to listen to these mixes many times?  Will this kind of production be part of a trend or will it stand the test of time?  Listen to Death Magnetic and answer these questions for yourself.
 
So what are my recommendations about loudness and compression?   Well for all of you that are musicians and sound engineers, it starts with the equipment.  I am a vintage gear user and a large number of my clients wanted to record at my studio because of my vintage gear collection.  I had numerous vintage compressors and limiters (both tube and solid state.)  These products are still available but at a hefty price.  The reason I prefer the vintage gear, and particularly tube compressors, is for what I call the “warmth” of the compression.   Another pivotal asset for tube compressors and limiters too is how each deals with a peak in the signal.
 
Tube distortion is, as you may recognize, that smooth and warm dirtying of the signal most frequently heard in guitar or bass amps.  It is not too different in tube compressors and limiters.  When the signal peaks it basically distorts, but with tube gear, the distortion can be an asset.  It adds “dirt” to the signal, but it does so with a smooth and warm result.  Digital distortion, which is the result of a signal peaking in the digital realm, is harsh, brittle and hard on the ears.  So, if you are going to use compression to get your recordings louder, I recommend spending as much as you can on high quality tube compressors and limiters.  I know it is expensive but the results are all the difference.  If this is not possible, and you work solely in the digital realm, do not sacrifice quality for loudness.
 
Metallica took a chance to win the loudness war, and they should be commended for trying to push the boundaries, but unless you have a budget like theirs, you will not be able to convince your listeners that the questionable sound in your mixes is worth it, even though it is so loud.

Logged

Jim Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 579

The "loudness war" is one where the listeners are the losers. The end result is crashing record sales, illegal downloading (why would anyone pay for that crap?) and increased book sales. (Silence is golden).

Now it's contaminated everything audio, TV was the last thing that got "regulated". The FCC told the broadcasters to even out the levels of their commercials, so they did. Now all their programming sounds like a Metallica record. Great.

Clipping, distortion and gratting mass compression is the "new world audio order". Like most things, it's been taken to the 9th degree and has ruined listening for me. I used to use an Aphex 612 expander to be able to listen to some of this crap, but with brick wall levels, there's no longer anything to grab onto.

So I prize my old Sony 1630 era CD's, remember those CD's that didn't go to "11"?

One thing I avoid are any "remastered" classic records. That's a warning to me that it will be messed up.

The real question on how all this got so bad is a basic one:

Why?
Logged

jaykadis

  • R/E/P Forums
  • Jr. Member
  • *
  • Offline Offline
  • Posts: 57


So I prize my old Sony 1630 era CD's, remember those CD's that didn't go to "11"?

That's a first - nostalgia for the 1630. It was an improvement on the 1610, though.

Jim Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 579

Compared to the recent "re-issues" I've heard, those 1630 CD's sound golden. They are also very relaxing to listen to. Yes, relaxing, remember when music used to do that?

Nearly all major label releases these days are ruined from mass compression. Like an insiduous disease, it's affected everything, even jazz.

It really doesn't matter how good their modern hardware is next to the 1630, the operators (squashing engineers) are doing a bang-up job of messing it all up, don't you think?

I do love live music, it's the only venue I can listen to these artists without the squashing engineer ruining it first. Thank god PA systems have limits to what their "limiters" can do.
Logged

Fletcher

  • Administrator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 590

The "Loudness Wars" have been fought since the Motown days... I remember Bob Ohlsson commenting on how they would reference Beatles recordings when he was a mastering engineer at Motown -- mainly for "loudness"!!

Missing a 1630 is kind of like missing a '72 Pinto... the memory is somehow much more fun than the reality [as always - YMMV].

Peace
Logged
CN Fletcher

mwagener wrote on Sat, 11 September 2004 14:33
We are selling emotions, there are no emotions in a grid


"Recording engineers are an arrogant bunch
If you've spent most of your life with a few thousand dollars worth of musicians in the studio, making a decision every second and a half... and you and  they are going to have to live with it for the rest of your lives, you'll get pretty arrogant too.  It takes a certain amount of balls to do that... something around three"
Malcolm Chisholm

Jim Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 579

It's wasn't the 1630, it was the operators. They didn't flat-line everything. Pull out any release from those days and compare, they are easier to listen to for extended periods. Current remasters are so squashed you get listening fatigue quickly. So you end up listening less, the law of unintended consenquences.

I don't blaim the arrows, it's the indians that are drunk on that compression whiskey. I just wish they would sober up.
Logged

Patrick Tracy

  • Newbie
  • *
  • Offline Offline
  • Posts: 34

If players (hardware or software) equalized the perceived average levels of recordings (downward in favor of those with higher peaks) there would be no incentive to squash the life out of music and the loudness war would be over in a day.
Pages: [1]   Go Up