The Loudness War

[Late]

From this page, which is worth reading:

In March 2010, mastering engineer Ian Shepherd organized the first "Dynamic Range Day," a day of online activity intended to raise awareness of the issue and promote the idea that "Dynamic music sounds better." The day was a modest success, and a follow-up has been announced on March 25th 2011, with industry support from companies like SSL, Bowers & Wilkins and Shure. Shepherd cites research showing there is no connection between sales and "loudness," and that people prefer more dynamic music. With music sales moving towards digital downloads and away from CDs, there is hope that the loudness war will be blunted by normalization technology such as Replay Gain and Apple's Sound Check.

... compression algorithms have been engineered specifically to accomplish the task of maximizing audio level in the digital stream. Hard limiting or clipping can result, affecting the tone and timbre of the music in a way that recording engineer Rip Rowan described as "sounds like dogshit." The effort to increase loudness has been referred to as the "Loudness Wars."

TOCJ 50086, The Fabulous Slide Hampton Quartet is the loudest jazz reissue I have ever heard. I think this session may have originally been recorded on the "hot" side, but for me this particular reissue is almost instantly fatiguing. A shame; it's one of my very favorite trombone-led quartet sessions. There are at least three different CD reissues of this album. I'll have to search others out.

Of the new "Jazz 999" TOCJ 500xx series, I've heard three titles so far. All are unusually loud. Even more of a shame; some titles are seeing their very first issue on compact disc. Maybe some other titles in the series will have more dynamic range than the ones I've heard.

It's depressing to me that the Japanese jazz reissue market has seemingly bought into the loudness war. I don't have an iPod or earbuds. I have nothing against them, but how much, in your opinion, are they a culprit in this "war"?

[Matthew]

"Loudness War" is pure bs cooked up by an "recording engineer" to boost his business.

[Chuck Nessa]

Loudness and compression are two different things. Be careful about confusing the two. Early cds were mastered at moderate volume levels (in fear of digital distortion) and things have changed. Excessive compression has become a relatively recent habit but they are not the same.

[Late]

Loudness and compression are two different things. Be careful about confusing the two.

I'm sure I confuse the two at times. In my understanding, I was under the impression that dynamic compression (or upward compression) results in a louder sounding recording, and that the term "loudness" refers to frequency response, which doesn't necessarily indicate that a recording itself will be, or is, "loud." The Wiki page I linked to at the top of the first post uses the term "loudness war," but that might be a misnomer, as the actual "war" appears to be in regard to compression and not frequency response. The loudness war, as a term and as an idea, doesn't seem (to me at least) like BS, but rather a standard practice — that compressed, and thus loud sounding, recordings (or television commercials, or radio broadcasts) capture attention through a loud "audio image," which is sometimes referred to as a wall of sound.

Maybe my real gripe is just with compression (at least as I understand it), and my surprise that it seems more and more common in Japanese jazz reissues. That's all — not trying to start a labored debate over compression and whether it's inherently bad.

[J.A.W.]

Excessively reducing (and sometimes virtually getting rid of) the dynamic range of recordings and making them louder, also known as the "loudness wars", seem to be common practice in pop/rock nowadays and it certainly isn't BS. Late is right, some Japanese jazz reissues also suffer from these practices.

[JSngry]

Does compression increase the actual decibel level of a sound or just its density?

[J.A.W.]

Does compression increase the actual decibel level of a sound or just its density?

It makes the quieter passages louder relative to the louder passages, i.e. the dynamic range of the music is reduced.

Excessive compression quashes the dynamic range.

[medjuck]

There are a couple of chapters on this and "clipping" in The Perfect Sound Forever. I didn't really understand it but apparently "" allows a song to sound louder when it comes on the radio.

[JSngry]

Does compression increase the actual decibel level of a sound or just its density?

It makes the quieter passages louder relative to the louder passages, i.e. the dynamic range of the music is reduced.

Excessive compression quashes the dynamic range.

Exactly - which is why compression doesn't make the music actually "louder". Loudness = volume = decibels.

What we're talking about with compression is not increased loudness, but decreased dynamic range, a flattening/condensing of the dynamic range, the relative volume, not the actual volume. It only seems louder because you're raising the floor, not lifting the ceiling.

Now, what usually results is a more prolonged/sustained period of higher-than-ordinary lower volumes. But the most loud portions of the sound stay right where they are.

You can say that a loudening of softer passages makes the music "louder", and maybe, but unless the peak volumes are bumped higher than they originally were, I think the distinction has to be made between compression and loudness, volume. Otherwise it's like saying that getting fatter makes you taller.

[J.A.W.]

Does compression increase the actual decibel level of a sound or just its density?

It makes the quieter passages louder relative to the louder passages, i.e. the dynamic range of the music is reduced.

Excessive compression quashes the dynamic range.

Exactly - which is why compression doesn't make the music actually "louder". Loudness = volume = decibels.

What we're talking about with compression is not increased loudness, but decreased dynamic range, a flattening/condensing of the dynamic range, the relative volume, not the actual volume. It only seems louder because you're raising the floor, not lifting the ceiling.

Now, what usually results is a more prolonged/sustained period of higher-than-ordinary lower volumes. But the most loud portions of the sound stay right where they are.

You can say that a loudening of softer passages makes the music "louder", and maybe, but unless the peak volumes are bumped higher than they originally were, I think the distinction has to be made between compression and loudness, volume. Otherwise it's like saying that getting fatter makes you taller.

What often happens in pop/rock nowadays is that both excessive compression is applied and the overall level is increased, so that not only the dynamic range is reduced or even quashed, but also the overall volume is maximized. Some Japanese and other jazz CD reissues also suffer from this phenomenon. An example is the track "Caravan" on the Lou Donaldson Mosaic (I sold that set years ago because of its sound; the Japanese "Blue Note Works" TOCJ versions of most of the Donaldson albums that were included in the set sound much better to my ears). Other examples are the Japanese Blue Note reissues of the late 1990s/early 2000s in the TOCJ-6xxx series. The ones I had sounded awful to my ears, loud and with hardly any dynamics.

[Daniel A]

Does compression increase the actual decibel level of a sound or just its density?

It makes the quieter passages louder relative to the louder passages, i.e. the dynamic range of the music is reduced.

Excessive compression quashes the dynamic range.

Exactly - which is why compression doesn't make the music actually "louder". Loudness = volume = decibels.

The way I understand it, the term 'loudness' does not equal decibels. Loudness is related to how a sound is perceived by the human ear, and compression typically makes a recording sound louder (to humans) when played at the same volume.

See Wikipedia: http://en.wikipedia.org/wiki/Loudness

Edited March 10, 2011 by Daniel A

[Larry Kart]

You can say that a loudening of softer passages makes the music "louder", and maybe, but unless the peak volumes are bumped higher than they originally were, I think the distinction has to be made between compression and loudness, volume. Otherwise it's like saying that getting fatter makes you taller.

I'm no expert here, but apropos Jim's introduction of "fatter" and "taller" -- if fatness refers only to what percentage of your body consists of fat, no, you wouldn't get any fatter if you got shorter, but you'd look fatter, i.e. there'd be more of you spreading sideways.

We're human beings, not machines; if compression makes things sound louder to us, then for us they are louder (though not to a device that measures volume). If you've got compression plus an increase in volume, the results will register as louder on a device that measures volume and will sound even louder than before to us.

[Late]

Exactly - which is why compression doesn't make the music actually "louder". Loudness = volume = decibels.

What we're talking about with compression is not increased loudness, but decreased dynamic range, a flattening/condensing of the dynamic range, the relative volume, not the actual volume. It only seems louder because you're raising the floor, not lifting the ceiling.

That makes perfect sense. I wonder, then, why I rush to the volume knob to turn a compressed recording waaay down. Is it only because I'm perceiving a "loud" sound?

When I play the Hampton reissue mentioned above, it's almost humorous how low I turn the volume knob. It's approximately at "2" of the 1-10 volume spectrum on my player.

[J.A.W.]

Exactly - which is why compression doesn't make the music actually "louder". Loudness = volume = decibels.

What we're talking about with compression is not increased loudness, but decreased dynamic range, a flattening/condensing of the dynamic range, the relative volume, not the actual volume. It only seems louder because you're raising the floor, not lifting the ceiling.

That makes perfect sense. I wonder, then, why I rush to the volume knob to turn a compressed recording waaay down. Is it only because I'm perceiving a "loud" sound?

When I play the Hampton reissue mentioned above, it's almost humorous how low I turn the volume knob. It's approximately at "2" of the 1-10 volume spectrum on my player.

You can't undo the effects of compression, what's gone is gone - in this case dynamics, quiet passages aren't really quiet anymore relative to the louder passages; turning the volume down doesn't really help, the "loudness threshold" will still be there.

As Jim described compression "sec" makes music seem louder, but, as I said earlier, it's often actually louder because there's not only compression but also an increase of the overall volume: what I would call "modern" masterings are often a combination of excessive compression and a maximum increase of the overall volume.

[Chuck Nessa]

I actually do this stuff and you have now confused me about the subject.

It is really simple - the compression of dynamics is a tool worth using sometimes and a tool worth ignoring other times. It is a tool. The use of the tool is what should be discussed. Simply branding something as "bad" or "good" is not helpful. We have this "blame game" stuff going on all the time and much of it is crap.

[J.A.W.]

I actually do this stuff and you have now confused me about the subject.

It is really simple - the compression of dynamics is a tool worth using sometimes and a tool worth ignoring other times. It is a tool. The use of the tool is what should be discussed. Simply branding something as "bad" or "good" is not helpful. We have this "blame game" stuff going on all the time and much of it is crap.

In his first post Late mentioned the problem he was having with the sound on a few Japanese CD reissues. As I understand it the problem he was talking about is excessive compression and maximized overall volume, not compression as such - which is indeed just a tool, I didn't brand it "bad" or "good". With my comments I tried to explain what I thought Late was hearing. If you can explain it better, be my guest; I'm sure it won't be crap

[.:.impossible]

Chuck, will you give an example of when it is worth it for an engineer to use compression?

From my limited exposure to this kind of thing, much of what has been written above, similes and metaphors etc., has completely obscured the subject.

I guess what I am asking is for you to put it in layman's terms without describing the layman.

[Chuck Nessa]

Simplest example - two track recording, the soloist wanders off mike. A take that might be otherwise discarded is saved by compressing the passage (with some gentle eq of the accompanying sounds) and the soloist seems to be in the proper perspective. It gets much more complicated with multi-tracks but drum kits can be rebalanced, etc. You can spend many hours doing this for the perfect master.

[.:.impossible]

Thank you.

[RDK]

"Loudness War" is pure bs cooked up by an "recording engineer" to boost his business.

It may be overstated by some but it's absolutely not bullshit. The problem is not making current CDs louder - it's the senseless reduction of dynamic range in order to achieve this that's the problem.

I don't put too much stock in charts and graphs when it comes to stuff like this, but the data here is interesting and mostly confirms the listening experience. http://www.dr.loudness-war.info/

Does compression increase the actual decibel level of a sound or just its density?

It makes the quieter passages louder relative to the louder passages, i.e. the dynamic range of the music is reduced.

Excessive compression quashes the dynamic range.

Exactly - which is why compression doesn't make the music actually "louder". Loudness = volume = decibels.

What we're talking about with compression is not increased loudness, but decreased dynamic range, a flattening/condensing of the dynamic range, the relative volume, not the actual volume. It only seems louder because you're raising the floor, not lifting the ceiling.

Now, what usually results is a more prolonged/sustained period of higher-than-ordinary lower volumes. But the most loud portions of the sound stay right where they are.

You can say that a loudening of softer passages makes the music "louder", and maybe, but unless the peak volumes are bumped higher than they originally were, I think the distinction has to be made between compression and loudness, volume. Otherwise it's like saying that getting fatter makes you taller.

It's the exact same effect you get on TV when a "loud" commercial comes on.

[RDK]

Chuck, will you give an example of when it is worth it for an engineer to use compression?

From my limited exposure to this kind of thing, much of what has been written above, similes and metaphors etc., has completely obscured the subject.

I guess what I am asking is for you to put it in layman's terms without describing the layman.

I'm not Chuck but probably 95% of all pop/rock music from the last 50 years has purposeful compression applied to it. It is a tool that's part of the production process, to get the right sound for a recording. The problem now is the prevalence of needless overcompression that's used to make a recording sound louder. It's no longer enough just to maximize the volume of a track, but to squash its dynamics to make it sound even louder.

[Chuck Nessa]

More than 50 years. That's why some folks complained their favorite 45s didn't sound right when transferred to lp or cd.

Correct about the current complaint.

[J.A.W.]

"Loudness War" is pure bs cooked up by an "recording engineer" to boost his business.

It may be overstated by some but it's absolutely not bullshit. The problem is not making current CDs louder - it's the senseless reduction of dynamic range in order to achieve this that's the problem.

I'm not Chuck but probably 95% of all pop/rock music from the last 50 years has purposeful compression applied to it. It is a tool that's part of the production process, to get the right sound for a recording. The problem now is the prevalence of needless overcompression that's used to make a recording sound louder. It's no longer enough just to maximize the volume of a track, but to squash its dynamics to make it sound even louder.

Thanks for explaining it much better than I did.

[Jazzmoose]

Just to make sure I'm understanding this discussion, Chuck is talking about tweaking a track or two, right? And the 'problem' recordings are what? They have the whole recording narrowed? If that's true, what is the stated reason for doing so?

[RDK]

Moose, check out the link I posted above. It's actually interesting to see the difference between older CD releases and some newer remasters. Check out some of the heavy metal albums especially - it's really pretty shocking. Fortunately, a lot of jazz and more acoustic music hasn't been affected... as much. I honestly don't hear it that much in the stuff I listen to, but it does seem pointless, and past a certain point it does adversely affect the sound quality.

This might help demonstrate the problem...