April 24, 2014

Why Dynamic Range is Just as Important in Your Audio as It is in Your Image

Michael Coleman is back with another SoundWorks Collection video, and this time it's the sound team from Wally Pfister's directorial debut Transcendence. The sound team talks a lot about the dynamic range in their audio mix, where they utilize very quiet sounds all the way up to very loud sounds in order to maximize the effectiveness of both. Check out the video below and learn why silence at the right moments can actually improve your sound mix.

The concept of dynamic range in audio is something that is talked about a lot in modern music, because most of the popular stuff you hear is mixed in a way that makes it sound louder than it really is. Essentially everything is mixed up, and there are no quiet sounds to balance it out. Because of the way movie theater speakers are designed, it's actually to the benefit of the creators not to do this, especially since you can direct where you want specific sounds to be in relation to the audience.

Using silence or quiet moments followed by intense sounds can be extremely powerful, and it was also something explored in the SoundWorks Collection video for All Is Lost:

Is the same thing that happened to music also happening with videos online? Since so many people watch videos on smartphones or laptops with puny speakers, is it a losing battle trying to build dynamic range into your sound mix? What do you think?

Links:

Your Comment

19 Comments

I think "Nine Inch Nails" reached the right solution with their latest album, "Hesitation Marks", which was released with 2 different mastering jobs: a conventional mix that is bassy, compressed to shit, and sounds loud, and an audiophile mix, which is much more dynamic.

Lots of people talk about loving dynamics in their music or movies, and when I am at home, with the headphones on in a dark room giving the media my full attention, that's exactly what I want, but there are other times when I'm listening to some tunes on the subway, and in that case, I want to hear the song, not just the loud parts.

There is a time and place for both, and having a compressed-for-loudness stereo track and a mixed-for-dynamics 5.1 track seems like the best of both worlds to me.

April 24, 2014 at 1:43AM, Edited September 4, 11:56AM

6
Reply

+1

April 25, 2014 at 1:35PM, Edited September 4, 11:56AM

0
Reply
Lcky

just another sad attempt at trying to make us watch that awful Transcendence movie.

April 24, 2014 at 2:26AM, Edited September 4, 11:56AM

10
Reply
joe bill

A movie can be awful and still use sound design effectively...

I think something that gets lost is that even with "bad" movies, when you get to this level, there are still a ton of craftspeople doing stellar work.

April 24, 2014 at 3:04AM, Edited September 4, 11:56AM

3
Reply
avatar
Joe Marine
Camera Department

I don't know about you guys but this look very helpful
I've been developing a technique to make sounds cinematic from my the camera audio nothing pluged in the camera and in the production magic happen it's not super great but it's enough to make it not video-ish

great article.

April 24, 2014 at 3:08AM, Edited September 4, 11:56AM

0
Reply
Hammam

The title of this article is "Why Dynamic Range is Just as Important in Your Audio as It is in Your Image" and I learned something. So what if there was a "Transcendence" trailer? The owners of this site have the right to make money, why should they work hard for nothing?

April 24, 2014 at 4:25AM, Edited September 4, 11:56AM

3
Reply
Raymond

One of the reasons, the modern music is "compressed to shit" because modern consumer audio can't reproduce the dynamic range - or anywhere close to it - of a live performance. Actually, the confined space of a movie theater is far more suitable to the high DR because one can achieve high volume without much distortion with the theater amplifiers and speakers. The second reason for the compression is that the modern music is heavily reliant on electronic effects and most popular singers are just piss poor at that singing part, requiring constant compression, auto tune, etc. On the other hand, if one likes largely acoustic music - from jazz to folk to serious opera to Broadway musicals - one could get very high fidelity sounds.
.
A move to Class D amplifiers also allows very high volumes with very little distortion ... of course, a Class A tube amp and some classy horns and that's as close to audio nirvana as one can get.

April 24, 2014 at 6:56AM, Edited September 4, 11:56AM

0
Reply
DLD

It's interesting that you compare Class A tube with audio nirvana, because that goes slightly against the audio DR discussion. I agree with you, vacuum tubes in Class A sound amazing, but that design is incapable of reproducing high dynamic range audio. The background noise of the tubes themselves will overshadow very quiet sounds, and anything loud will overdrive the tubes, losing clarity.
I, for one, sometimes loathe the use of loud and soft sounds next to each other, because sometimes it's just painfully jarring. In the natural world, there is rarely an instance where noise goes from 50db to 110db instantly, there is usually a build-up to that higher volume. I think that's what should be desired in film as well. Having that build up can be suspenseful, all while avoiding the jarring nature of an audio jump-cut.
For the record, I also didn't like the sudden dark-to-light cuts in Castaway, so if you do enjoy that sort of thing, feel free to disagree with me.

April 24, 2014 at 9:23AM, Edited September 4, 11:56AM

0
Reply
Joel

@ Joel - this is why I mentioned horns, which let Class A tube breathe without taxing the tubes for volume. Tubes actually work fine during the audible range upward but they have problems delivering bass (in Class A). That is why, for the best buys, powered subs are pretty much solid state and/or Class D.
.
@ Alex - I said nothing about the loudness wars per se. And, of course, you can make the over-processed Auto-Tuned voice louder than its accompaniment but that's hardly a good recording or mixing technique.
.
@ tcal - you can actually pick up a tube amp from the 1920's - a restore project, to be sure, but there are some - and hear a more pleasant reproduction of sound than a modern heavily compressed format. Once again, I said nothing about the boom-boxes or the Walkman. You said nothing about the modern ultra high end tube amplifiers like Lamm (top of the line amp is $140,000) or Audio Research ($55,000 for a pair or mono blocks).

April 24, 2014 at 6:33PM, Edited September 4, 11:56AM

4
Reply
DLD

DLD, you don't exactly know what you're talking about. Compression of an overly-dynamic vocalist in a music track is not at all the same as the problem of the loudness wars in music. And there is nothing inherently loudness-induced in electronic effects. The problem is crushing/shedding the overall peaks of a track, typically at the mastering stage. to the point where it's a mushy, distorted, fatiguing mess just so its playback level will seem louder than other tracks. There is plenty of room for all types of creative compression, as compressors are just additional colors in the creative palette, but the real issue here has nothing to do with that and everything to do with senseless competition that does nothing more than destroy the fidelity of music.

April 24, 2014 at 2:03PM, Edited September 4, 11:56AM

1
Reply
Alex

DLD u are one of those old heads full of shit. Everything old isn't gold. Those good old days are mostly based on youthful memories. Before the mp3 generation, the sound coming out those crappy headphones from the discman in the 90's and walkmans of the 80's and boomboxes of the 70's was pure garbage to begin with, therefore, the youth in those days didn't have a better music listening experience than today.
Secondly, u do realize the noise generated by those tube compressors u prefer. Compression in general and loudness generated during mastering serve two different purposes

April 24, 2014 at 3:04PM, Edited September 4, 11:56AM

1
Reply
t.cal

Are you kidding, t.cal?!

Apart from lossless audio files like .FLAC, a CD is the best quality digital recording you can hear. Generally, a .WAV or .AIFF file is 10MB/sec whereas an .mp3 is only ~1MB/sec. Mp3s have 1/10th the audio fidelity of CDs. Where do you get your information?

Here, read up: http://www.stereophile.com/features/308mp3cd

April 24, 2014 at 4:15PM, Edited September 4, 11:56AM

2
Reply
Nick

No. you must be kidding to think that every cd being listened to was .wav format and most importantly, that the MEDIUM BY WHICH THE CD WAS BEING LISTENED TO AND NOT ACQUIRED FILE - the headphones being sold with those cd players - were capable of producing the quality of sound coming out the vd players. Was your comment a joke? What was the purpose?

April 25, 2014 at 7:53AM, Edited September 4, 11:56AM

4
Reply
t. cal

Nothing beats McIntosh amplifiers.

April 26, 2014 at 1:27AM, Edited September 4, 11:56AM

0
Reply
Richard

The thing about dynamic range is that you don't need theater-quality sound systems to exploit and use it well. I used to mix student films at USC on 16mm mag (mono, mind you), the dynamic range available to me was freaking terrible, but we were still able to create quiet moments, loud moments, use sound design for pacing and mood in ways beyond just loud loud loud. Yes there was less room to play around in, but as long as you're thinking creatively about having quiet moments, having loud moments, and avoiding wall-to-wall music you're in a good spot.

A lack of dynamic range doesn't mean you can't have a dynamic mix.

April 24, 2014 at 10:26AM, Edited September 4, 11:56AM

1
Reply

I love this series, but it would be so much more helpful if they profiled a couple low/no budge films as well.

April 25, 2014 at 1:00AM, Edited September 4, 11:56AM

0
Reply
Justin

It's not low budget or indie, but "Bring it On" has FANTASTIC sound design for what it is. IMO. Nicely done, subtle, and thematic.

April 25, 2014 at 1:15PM, Edited September 4, 11:56AM

2
Reply

This is a huge, massive problem. MOST and studio monitors do not come even close to a theaters dynamic range. Your usable range in a theater is between -20db to -6db in most audio mixing systems. Where at best most studio monitors give you -8db to 0db. You cannot mix to 0db peaks, you will blow up the theater aka blow out their speakers.

Sadly the only way to get this type of fix is to have the correct speakers. Studio monitors don't cut it. I have mixed sound in 5.1 and delivered it in DCP form to be played in the second largest house at the Muvico Theaters in Rosemont IL. You have to know where to work also, but if you don't have a setup to hear the range that the theater produces at reference, you'll always be off.

For home theater you want to push things up and compress it a bit but not for theatrical. We used JBL LSR monitors and Adobe Audition, We are switching out setup over to match more theatrical sound using speakers that are used in the theater.

April 25, 2014 at 12:07PM, Edited September 4, 11:56AM

7
Reply
Darren Orange

I've recorded albums on 2" 24Track (back when budgets allowed for it) through automated entirely analogue desks (I had a Euphonics 3000CS). Beautiful. Then I'd track and mix feature films on multiple 16bit Protools systems for Dolby encoding down to MP2 audio files for cinema release. Not great, when you've heard REAL audio... but like the film vs digital video debate, it's all budget vs fidelity vs time.

Dynamic range... mixing for TV, we have to adhere to very stringent dynamic ranges (or else the material won't get thru tech check for broadcast). For features, same thing, but you get more DR. For music, anything goes pretty much, although there are generally accepted ranges. I guess I'm saying there are lots of element that go into the mix. It's almost easier these days shooting rather than mixing! ;)

April 25, 2014 at 8:23PM, Edited September 4, 11:56AM

0
Reply

The problem is not that the vast majority of audio speakers have limited range. The problem is that audio mixers no longer control the levels. What works in the first weeks of a film in the theater does not work anywhere else. With so many layers of sounds, very often the dialog track is lower than the sound effects. If a viewer has to touch the volume buttons between loud and really quiet scenes, they are drawn out of the film.

May 19, 2015 at 2:09AM

0
Reply
Brandon
81