• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Dynamic range, loudness war, remasters.

skymusic20

Member
Joined
Aug 19, 2020
Messages
47
Likes
16
Hi all!
Im confused about the so called loudness war.

First, I wonder why DR compression became a thing with CDs. It seems to me nobody was discussing loudness war before CD. But I might be wrong. Yes, there was DR compression but I don't remember it was an issue.

I have read that making music sound louder makes people wrongly think it sounds better, like louder=better, hence people buy it. But by the time CDs arrived, people were already enthusiastically buying DR uncompressed vinyl. So from the pure marketing point of view, I see no reason to DR compress music. There was no need to convice people to buy CDs. CDs where the real deal back then and everybody loved them. Then, what was the need to DR compress music when CD arrived? The argument that this was done purely for marketing reasons does not convince me. But Im no marketing expert and I might well be completely wrong.

My second question is perhaps a heresy. Can a DR compressed album (remastered) sound better (be more faithful) than the previously DR uncompressed version of it?

Let me explain. Lets take any Led Zeppelin album. It seems to me that the newer the version, the less dynamic range it has. There have been countless of versions of Led Zeppelin IV, for instance. As you go from vinyl to CD and as years go by, every newer version seem to have less DR. As far as I know, band members have been involved in the remasters. They seem to endorse all remasters. The same happens with Iron Maiden remasters. As far as I know, band members seem to claim their newer versions (remasters) sound better. Yet these newer versions have less DR. But can you argue against the artist? Fidelity seems to be related to being faithful to what the artist intended. If artists tell you their remasters sound better, does it make sense to argue against the artist?

Now, Im probably gonna say something very, very foolish because I know my knowledge is very limited (Im basically ignorant and I admit it): Is it possible that one particular album shows a very wide DR but in truth some of that wide DR is due to some very, very soft noise introduced in the recording process? I mean, very soft sounds or noise that are very soft but that should not be there, not part of the song, but somehow got their way into the final recording.... Then if you release a new version and you cut out such unwanted soft noise, DR might seem to be reduced when in fact, it is not.

Now, Im also confused because some people in music forums strongly (almost religiously) recommend to buy japanese remasters of old CDs. For instance, I have read people who absolutely recommend buying the japanese remasters of Judas Priest. But when I look for them, these japanese remasters are also heavily compressed. The same people consider the UK/US 2001 Judas Priest remasters to be absolute crap, but the newer Japanese remasters to be absolute gold. Yet the japanese remasters are heavily comporessed. So, why one must be better than the other? Why one compressed version is garbage while the other is diamond if both are DR compressed?

Finally, when getting into softer more instrumental music, I notice a trend. For instance, Mike Oldfield and Alan Parsons remasters seem to preserve more or less their original DR found in the vinyl. Still, CD versions are more compressed than vinyl versions but not much. In general terms remastered CDs from these artists have a fairly wide DR but never as much as vinyl.

As to my personal opinion, I like almost any CD version, compressed or not from my favourite albums. I can not say, for instance, that Procol Harum first CDs sound better or worse than new remasters. Nor I can say that Judas Priest 2001 remasters are absoulte crap. Yes, they sound louder but that's not necessarily bad. Sometimes they are brighter but not much.

Finally, when CD versions were released, I remember being somewhat disapointed about a few albums: Mike Oldfield Five Miles Out, Deep Purple In Rock, UFO Phenomenon, Bob Welch Three Hearts. They sound a bit weak. Perhaps the hype of the CD was so big that I expected too much. And when I finally bought them I was like oh, where is all the energy? But that was perhaps psychologically. Because I expected a whole new audio experience. And these particular albums were to my ears a bit lacking in "energy" whatever that means. Yet they are some of my absoulte favorite albums and still love them as they are, vinyl, CD, CD remastered or whatever.

Here I have talked about rock, pop and metal and sometimes I think for such kind of music DR compression is unimportant.

As far as I know, classical music CD are never DR compressed and if done it would be a huge mistake, I guess.

But I wonder, is it a mistake to buy remasters? Or is it a case by case situation?

Are all remasters crap? Or are they any good as long as they keep a wide DR. If so, how much?

Thanks all.
 

Weeb Labs

Senior Member
Forum Donor
Joined
Jun 24, 2020
Messages
372
Likes
814
Location
Ireland
Hi all!
Im confused about the so called loudness war.

First, I wonder why DR compression became a thing with CDs. It seems to me nobody was discussing loudness war before CD. But I might be wrong. Yes, there was DR compression but I don't remember it was an issue.

The loudness wars preceded CDs by approximately 40 years. The perceived volume of masters was quite important at the time, as louder recordings were more likely catch the attention of radio listeners. When mastering for vinyl, the medium's inherent physical characteristics limit the amount of compression that can be applied. Digital media such as CD (and certain analog media) introduce no such limitations and so studios were afforded the freedom to make their masters as loud as desired.

My second question is perhaps a heresy. Can a DR compressed album (remastered) sound better (be more faithful) than the previously DR uncompressed version of it?

This is rather subjective but there are many remasters possessing less dynamic range than the originals which most people find preferable; AC/DC's Shoot to Thrill being one such example. More dynamic range is not always preferable.

Now, Im probably gonna say something very, very foolish because I know my knowledge is very limited (Im basically ignorant and I admit it): Is it possible that one particular album shows a very wide DR but in truth some of that wide DR is due to some very, very soft noise introduced in the recording process?
Not by any appreciate amount. If the noise floor of a given recording is audible, this would in fact reduce the maximum dynamic range.

So, why one must be better than the other? Why one compressed version is garbage while the other is diamond if both are DR compressed?

As mentioned, more dynamic range is not always preferable. Depending upon the individual recording and the genre of music, it can absolutely be preferable to most listeners (audiophiles included).

But I wonder, is it a mistake to buy remasters? Or is it a case by case situation? Are all remasters crap? Or are they any good as long as they keep a wide DR. If so, how much?
If you are at all familiar with Queen's earlier albums, then I'm sure you will agree that the remasters are a significant improvement! :p

To most listeners, anywhere from 8dB to 15dB of average dynamic range will typically sound quite pleasant but once again, it is highly dependent upon the content. Some modern music is crafted with limited dynamic range being an intentional component. The trouble often arises when a studio decides to further compress music which was never intended to be "loud", destroying the nuance of the recording.
 
Last edited:

DVDdoug

Addicted to Fun and Learning
Joined
May 27, 2021
Messages
549
Likes
680
First, I wonder why DR compression became a thing with CDs. It seems to me nobody was discussing loudness war before CD. But I might be wrong. Yes, there was DR compression but I don't remember it was an issue.
As far as I know it compression wasn't viewed as a "problem". From what I've read, Motown was "winning" the loudness war with a process they called "Loud And Clear".

The big thing that's changed since the analog days (and the early digital days) is digital compression which allows tons of compression with look-ahead, and multi-band compression, etc., and they can do this with a minimum of artifacts/side effects.

When CDs were introduced, I anticipated that music (style and production ) would become more dynamic to take advantage of the wider dynamic range... Boy, was I wrong!!!

I have read that making music sound louder makes people wrongly think it sounds better, like louder=better, hence people buy it.
"Better" is a matter of personal opinion and taste so we can't really say "wrongly". I suspect that louder (and more compressed) makes a better first-impression but a more dynamic recording MIGHT be more enjoyable (and less boring) with longer listening or multiple listenings. I find a lot of modern music boring but I don't know if I'd like it more if it were less compressed.... Maybe it's just because I'm old and I like my "good old stuff".

But by the time CDs arrived, people were already enthusiastically buying DR uncompressed vinyl. So from the pure marketing point of view, I see no reason to DR compress music. There was no need to convice people to buy CDs. CDs where the real deal back then and everybody loved them. Then, what was the need to DR compress music when CD arrived? The argument that this was done purely for marketing reasons does not convince me. But Im no marketing expert and I might well be completely wrong.
Marketing is EVERYTHING! It's called the record BUSINESS. On vinyl or CD it needed air-play to become a hit. Radio stations have their own compression, but that's added to whatever is on the recording.

My second question is perhaps a heresy. Can a DR compressed album (remastered) sound better (be more faithful) than the previously DR uncompressed version of it?
I assume the people making the decisions (and most consumers) consider it "better". "More faithful" doesn't have much meaning with rock/popular studio recordings.

Let me explain. Lets take any Led Zeppelin album. It seems to me that the newer the version, the less dynamic range it has. There have been countless of versions of Led Zeppelin IV, for instance. As you go from vinyl to CD and as years go by, every newer version seem to have less DR. As far as I know, band members have been involved in the remasters. They seem to endorse all remasters. The same happens with Iron Maiden remasters. As far as I know, band members seem to claim their newer versions (remasters) sound better. Yet these newer versions have less DR. But can you argue against the artist? Fidelity seems to be related to being faithful to what the artist intended. If artists tell you their remasters sound better, does it make sense to argue against the artist?
The artist rarely has any say in it. ;) And the artists are just as likely as anybody else to fall-into the "louder is better" trap. And, they have every incentive to push sales of their newly-remastered music.

Now, Im probably gonna say something very, very foolish because I know my knowledge is very limited (Im basically ignorant and I admit it): Is it possible that one particular album shows a very wide DR but in truth some of that wide DR is due to some very, very soft noise introduced in the recording process?
No. Reducing the noise actually makes a wider dynamic range (or "dynamic contrast") depending on how you measure it. But, with most measurements/calculations it would have very little effect.

...The vinyl cutting and playback process (including the RIAA recording & playback EQ) changes the wave shape and introduces phase shifts. That affects the way the peaks line-up making some peaks higher and some peaks lower. A similar thing happens when you make an MP3. That doesn't affect the sound of the dynamics, but if you measure the crest factor (peak to average ratio) the vinyl will "measure" more dynamic than the CD (made from the same master) because of the new higher-peaks. A lot of people are fooled into thinking the vinyl has wider dynamic range because of those measurements. And, blind listening isn't that helpful because you can always tell the vinyl (from the noise).

I tried an experiment with MP3 vs the original CD, checking the crest factor and EBU R128 Loudness Range. (I haven't tried this with vinyl because I don't have much vinyl anymore and I have no idea if the CD & vinyl were differently mastered.) As expected, the MP3 has a slightly higher crest factor but the Loudness Range was the same.

Now, Im also confused because some people in music forums strongly (almost religiously) recommend to buy japanese remasters of old CDs. For instance, I have read people who absolutely recommend buying the japanese remasters of Judas Priest. But when I look for them, these japanese remasters are also heavily compressed. The same people consider the UK/US 2001 Judas Priest remasters to be absolute crap, but the newer Japanese remasters to be absolute gold. Yet the japanese remasters are heavily comporessed. So, why one must be better than the other? Why one compressed version is garbage while the other is diamond if both are DR compressed?
Different mastering engineers with different tastes? Somewhere (maybe on this forum) I read that Japanese mastering engineers work in "dead" rooms whereas most American mastering engineers work in rooms with more reflections...

They sound a bit weak. Perhaps the hype of the CD was so big that I expected too much. And when I finally bought them I was like oh, where is all the energy? But that was perhaps psychologically. Because I expected a whole new audio experience. And these particular albums were to my ears a bit lacking in "energy"
A louder, more-compressed, recording is "more intense". "Intensity" is probably a better word because we all have a volume control and the the engineers/producers can't really control the loudness. Some people will actually describe the more compressed recording as "more dynamic".

Here I have talked about rock, pop and metal and sometimes I think for such kind of music DR compression is unimportant.
It's just a matter of taste. Dark Side Of The Moon seems dynamic (compared to most rock recordings) but I don't think it's as dynamic as classical or jazz... I do enjoy the dynamics of some rock recordings... I like to hear the drums & cymbals "pop out" instead if being mushed with the rest of the music. To me, I think it's the short-term dynamics that are important.

I don't find myself adjusting the volume control in the car with Pink Floyd and I don't normally listen to classical, But, I have some Broadway soundtracks where I have to turn-up the quiet parts in the car and then turn it back down when the loud parts come-back.

As far as I know, classical music CD are never DR compressed and if done it would be a huge mistake, I guess.
I think they usually use SOME compression. I have heard of cases where there was none. Most people don't listen to classical music (or rock music) at "realistic levels" so with the full dynamic range the quiet parts would be too quiet when listening at moderate levels.

I read about a remix done for a rock band's box set... I'm thinking Journey but ATM that doesn't seem like the right band... They CLAIMED they weren't using ANY compression but then later they mentioned saturating the mixing console (it was re-mixed on an analog console but the tapes were digitized before they started). I don't remember if they said anything about the mastering...

But I wonder, is it a mistake to buy remasters? Or is it a case by case situation?

Are all remasters crap? Or are they any good as long as they keep a wide DR. If so, how much?
Yeah, it's case-by-case. I read another story about remastering (and remixing?) Led Zeppelin and they said the bass (especially the kick drum) was rolled-off in the original master and it was retained on the re-master. So, maybe the latest remaster is "better"? (I didn't buy it.... I have a Greatest Hits collection but I'm not a huge Led Zeppelin fan.)
 

gsp1971

Senior Member
Forum Donor
Joined
May 26, 2021
Messages
383
Likes
588
Location
Athens, Greece
Hi skymusic20,

Was in your shoes some 5 years ago, ended up throwing to the garbage most of my CD collection and I started over.
So I will try to provide some answers here.

First, I wonder why DR compression became a thing with CDs. It seems to me nobody was discussing loudness war before CD. But I might be wrong. Yes, there was DR compression but I don't remember it was an issue.
DR compression existed before the CD as well. It is just that with vinyl there is a limit to how much DR compression can occur before the stylus physically starts jumping out of the grooves. With CD and, later, digital downloads, it has gotten worse. As @Weeb Labs said, it started out as a means to try and capture the attention of radio listeners. It has gotten worse over the years, as young people use their phones to listen to music through bad quality earphones, hence they made the songs louder.

I have read that making music sound louder makes people wrongly think it sounds better, like louder=better, hence people buy it.
True, many people will confuse 'louder' with 'better'.

My second question is perhaps a heresy. Can a DR compressed album (remastered) sound better (be more faithful) than the previously DR uncompressed version of it?
First of all, let me say that decreasing DR from a recording is non-reversible, you can never get it back. In that sense, more DR in a recording is usually better. Having said that, there can be cases when someone remasters an album from the original analog tapes and produces a result with lower (or higher) DR than the equivalent LP / CD release of the time. Losing a point in DR (say from 13 to 12) is not the end of the world. If the remaster has been done with care, the remastered album can sound better. Examples include remasters of old albums by Analogue Productions and Mobile Fidelity Sound Lab (MFSL).

Let me explain. Lets take any Led Zeppelin album. It seems to me that the newer the version, the less dynamic range it has. There have been countless of versions of Led Zeppelin IV, for instance. As you go from vinyl to CD and as years go by, every newer version seem to have less DR. As far as I know, band members have been involved in the remasters. They seem to endorse all remasters. The same happens with Iron Maiden remasters. As far as I know, band members seem to claim their newer versions (remasters) sound better. Yet these newer versions have less DR. But can you argue against the artist? Fidelity seems to be related to being faithful to what the artist intended. If artists tell you their remasters sound better, does it make sense to argue against the artist?
This is exactly what the Loudness War is. Every time you reduce DR, you actually uniformly make the entire track louder by say, 3dB. But parts of the track that were already loud in the original recording, now that they are amplified by 3dB actually surpass the maximum 0dB possible loudness, and since this is not possible, these parts of the track get 'clipped', cut-off, you lose the musical information.
Example: a Led Zeppelin track has its softest sound at -13dB and its louder sound at -1dB for a dynamic range of 12. If I want to make the track louder and increase everything by 3dB, the -13dB now becomes -10dB and the -1dB should become +2dB, but since 0dB is the maximum, whatever musical info does not fit, it gets clipped.
Many people still chase releases from the 80s and 90s for that very reason.

Now, Im also confused because some people in music forums strongly (almost religiously) recommend to buy japanese remasters of old CDs. For instance, I have read people who absolutely recommend buying the japanese remasters of Judas Priest. But when I look for them, these japanese remasters are also heavily compressed. The same people consider the UK/US 2001 Judas Priest remasters to be absolute crap, but the newer Japanese remasters to be absolute gold. Yet the japanese remasters are heavily comporessed. So, why one must be better than the other? Why one compressed version is garbage while the other is diamond if both are DR compressed?
It is true that Japanese remasters have a good reputation. But not all are good, just as not all non-Japanese remasters are bad. People who follow this things (myself included) tend to trust certain remastering companies and engineers rather than country of origin.
Use this link to find high DR releases of the music that interests you.
https://dr.loudness-war.info/


Finally, when getting into softer more instrumental music, I notice a trend. For instance, Mike Oldfield and Alan Parsons remasters seem to preserve more or less their original DR found in the vinyl. Still, CD versions are more compressed than vinyl versions but not much. In general terms remastered CDs from these artists have a fairly wide DR but never as much as vinyl.
Some genres of music tend to suffer more from compression than others. Jazz, classical and instrumental fare better than pop, rock, and hip-hop.

As to my personal opinion, I like almost any CD version, compressed or not from my favourite albums. I can not say, for instance, that Procol Harum first CDs sound better or worse than new remasters. Nor I can say that Judas Priest 2001 remasters are absoulte crap. Yes, they sound louder but that's not necessarily bad. Sometimes they are brighter but not much.
Compressed DR has been proven to have a strong correlation with listening fatigue. This extra loudness / brightness will eventually result in listening fatigue, especially if your speakers are on the bright side.

Finally, when CD versions were released, I remember being somewhat disappointed about a few albums: Mike Oldfield Five Miles Out, Deep Purple In Rock, UFO Phenomenon, Bob Welch Three Hearts. They sound a bit weak. Perhaps the hype of the CD was so big that I expected too much. And when I finally bought them I was like oh, where is all the energy? But that was perhaps psychologically. Because I expected a whole new audio experience. And these particular albums were to my ears a bit lacking in "energy" whatever that means. Yet they are some of my absolute favorite albums and still love them as they are, vinyl, CD, CD remastered or whatever.
It depends on the release. Use the link above.

Here I have talked about rock, pop and metal and sometimes I think for such kind of music DR compression is unimportant.
No it is not. Play a compressed release of an Eagles or Dire Straits album and an uncompressed release of the same album and you will see the difference.

As far as I know, classical music CD are never DR compressed and if done it would be a huge mistake, I guess.
They can be compressed, but usually mildly or not at all because such music is targeted to a different audience. Most classical music lovers will probably listen through a proper hi-fi system rather than on crappy earphones on the subway.

But I wonder, is it a mistake to buy remasters? Or is it a case by case situation?
It depends on the release. Use the link above.

Are all remasters crap? Or are they any good as long as they keep a wide DR. If so, how much?
It depends on the release. Use the link above.

Through the above link, I have found great releases for my favorite albums but I have also discovered some wonderful uncompressed music.

A final word. Whether CD or Vinyl, when you listen to a very well recorded album on a good set of speakers, the experience is so spine-tingling that you'll never want to go back to heavily compressed stuff.
There are labels which produce good quality recordings, such as ECM, Fone, ACT, Chesky, Reference Recordings, etc. Have a look at them, see if they have anything you like and give it a try and tell me whether you liked it.

Hope I could help.
 
Last edited:

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
1,173
Likes
1,674
Location
San Diego
For a lot of my favorite older music I have the original LP, original CD, and sometimes a re-mastered CD / SACD/ download if it has a good reputation. I also have streaming so I can hear what ever the latest re-master sounds like. I always enjoy the original LP as a reference to what the "original" sounded like (original sound NOT necessarily the best / preferred sound) but beyond that I would say it is completely random as to what version "sounds better / I prefer"..... it really is a case by case basis with no rhyme or reason. If I had to pick the safest bet I would say "early CD's" as I prefer more dynamic mastering and many (but certainly not all) original CD's are both very good and very cheap. Not sure what advice to give you besides get multiple versions of some music you like and see what version you prefer and then try to figure out why (FFT analysis in Adobe Audition is where I start). Good luck and have fun.
 

antennaguru

Senior Member
Joined
Jun 16, 2021
Messages
321
Likes
261
Location
USA
The absolute best Led Zeppelin recording I have ever heard is the DVD-Audio disc "How the West Was Won". It is a live performance and very well engineered. The more quiet and delicate Zep songs that I really enjoy do indeed get quiet enough that the sounds of the audience become immersive. When I get goose-bumps I know it's a good recording!
 

j_j

Major Contributor
Audio Luminary
Technical Expert
Joined
Oct 10, 2017
Messages
1,282
Likes
2,568
Location
My office, custom listening room, or anywhere.
As several people have pointed out, you can't compress the (*(*&(* out of an LP the way you can out of a CD. The mechanics of the LP won't permit it.


So, when the ability to MAKE IT EVEN LOUDER was given to the A&R guy, there y'go.
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,262
Likes
2,412
Location
Pacific Northwest
...
As far as I know, classical music CD are never DR compressed and if done it would be a huge mistake, I guess.
They can be compressed, but usually mildly or not at all because such music is targeted to a different audience. Most classical music lovers will probably listen through a proper hi-fi system rather than on crappy earphones on the subway. ...
That's mostly true. Here's an example of an exception that proves this rule: https://play.primephonic.com/album/3760195734520
So disappointing, because that's a great performance ruined by excessive dynamic compression.
Other than these rare examples, classical music is blessedly outside the blast radius of the loudness wars.
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,262
Likes
2,412
Location
Pacific Northwest
...
This is exactly what the Loudness War is. Every time you reduce DR, you actually uniformly make the entire track louder by say, 3dB. But parts of the track that were already loud in the original recording, now that they are amplified by 3dB actually surpass the maximum 0dB possible loudness, and since this is not possible, these parts of the track get 'clipped', cut-off, you lose the musical information. ...
That's the general idea, but it's not quite that simple. They don't simply make the entire track louder, but instead use dynamic compression algorithms that boost the quiet parts a lot, boost the louder parts only a little, and use soft limiting to avoid clipping. So the entire track can be made maximally loud without clipping or loss of information. This still squashes the life out of the music, as dynamic range is an entire dimension of musical expression that is destroyed in this process. And it makes music sound unnatural, as loudness is part of timbre and detail perception.
 

gsp1971

Senior Member
Forum Donor
Joined
May 26, 2021
Messages
383
Likes
588
Location
Athens, Greece
That's the general idea, but it's not quite that simple. They don't simply make the entire track louder, but instead use dynamic compression algorithms that boost the quiet parts a lot, boost the louder parts only a little, and use soft limiting to avoid clipping. So the entire track can be made maximally loud without clipping or loss of information. This still squashes the life out of the music, as dynamic range is an entire dimension of musical expression that is destroyed in this process. And it makes music sound unnatural, as loudness is part of timbre and detail perception.

Correct. Although I am not an expert by any means, I have researched a lot on the loudness wars re: how mastering engineers use algorithms, limiters, etc. They don't necessarily clip information but make it all flat at the 0dB point, which as you said, squashes the life out of the music and makes it unnatural.
 

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
1,173
Likes
1,674
Location
San Diego
That's the general idea, but it's not quite that simple. They don't simply make the entire track louder, but instead use dynamic compression algorithms that boost the quiet parts a lot, boost the louder parts only a little, and use soft limiting to avoid clipping. So the entire track can be made maximally loud without clipping or loss of information. This still squashes the life out of the music, as dynamic range is an entire dimension of musical expression that is destroyed in this process. And it makes music sound unnatural, as loudness is part of timbre and detail perception.

I compare a lot of different mastering's in Foobar ABX and Adobe Audition visually and using FFT to compare the FR response (which for me corresponds to me being able to ABX different mastering's more than any other factor). My question is do these active compression technics affect the FR as seen in a FFT? I see "timbre" and "detail" mentioned as being affected by compression but aren't these changes just altered FR or something else?
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,262
Likes
2,412
Location
Pacific Northwest
I compare a lot of different mastering's in Foobar ABX and Adobe Audition visually and using FFT to compare the FR response (which for me corresponds to me being able to ABX different mastering's more than any other factor). My question is do these active compression technics affect the FR as seen in a FFT? I see "timbre" and "detail" mentioned as being affected by compression but aren't these changes just altered FR or something else?
Our perception of frequency response depends on loudness. That is, as you turn the volume up and down, we perceive a change not only in loudness, but also in frequency response. So when dynamic compression lifts the quiet parts, they don't just get louder, it also changes their timbre. This is evident in equal loudness contours; the perceived frequency response curve flattens as it gets louder.
Detail is similarly affected by loudness. Subtle detail that is down in the mix, like musicians breathing or clothes rustling, gets blown up out of proportion, especially in the quiet parts.

PS: to address your question directly: if you boost a quiet section by some constant amount, the FFT won't change. You haven't changed the frequency response. But, being louder, you will perceive/hear it differently, as if you had boosted the bass & treble. You can test this for yourself by simply turning up the volume.

PPS: This reminds me of a comment in the liner notes of Jakob Lindberg's performances of John Dowland; on p. 35 he says, "The listener is kindly requested to listen at a reduced volume when playing lute recordings. Otherwise all the nuances disappear and the resulting timbre is most incorrect." [emphasis in original]
 
Last edited:

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
1,173
Likes
1,674
Location
San Diego
PS: to address your question directly: if you boost a quiet section by some constant amount, the FFT won't change. You haven't changed the frequency response. But, being louder, you will perceive/hear it differently, as if you had boosted the bass & treble. You can test this for yourself by simply turning up the volume.

When comparing mastering's I take the entire song and in Audition and do a "Frequency Analysis" which is really an FFT but of the entire song. It seems to me if the "boosted" part of the music has different frequencies compared to the loud parts then the FFT would change? See below, this is 3 versions of a song, LP, Original CD, and Re-mastered download. I level matched them to the same LUFS and ran the frequency analysis tool. The "remastered" (red) is quite compressed compared the original CD (yellow) or LP (purple) versions and like almost all "compressed" music I look at has higher energy from 2 Khz to 9 Khz compared to the uncompressed version. I thought this was EQ added during the re-mastering but now I am thinking this is just what happens with heavy compression. In any case I can certainly hear and ABX this as increased "brightness" which is not appreciated, especially for vocals.

I will never marry red remstered yellow CD purple LP.jpg
 

dc655321

Major Contributor
Joined
Mar 4, 2018
Messages
1,091
Likes
1,391
PS: to address your question directly: if you boost a quiet section by some constant amount, the FFT won't change. You haven't changed the frequency response. But, being louder, you will perceive/hear it differently, as if you had boosted the bass & treble. You can test this for yourself by simply turning up the volume.

How could one alter the signal in the time domain (compression) and leave the frequency domain unaffected?
The spectrum must change too.

EDIT: ah, missed the "constant amount". That's sort of a misleading, special case isn't it?
Compression is not constant.
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,262
Likes
2,412
Location
Pacific Northwest
When comparing mastering's I take the entire song and in Audition and do a "Frequency Analysis" which is really an FFT but of the entire song. It seems to me if the "boosted" part of the music has different frequencies compared to the loud parts then the FFT would change? ...
The FFT will change only if they EQed it. The FFT will not change if they simply boost the level of quiet parts and do nothing else. You can test this yourself. Take the FFT of a quiet section of music. Boost it 12 dB and take the FFT again. The FFTs will be identical. By this I mean identical in shape / frequency content. Of course the amplitudes of the boosted one will be higher. But they'll be higher equally across all frequencies so the shape won't change.

However, even though the frequency response is the same, we won't perceive it that way. The boosted one, being louder, will sound like it has been EQed to boost bass & treble relative to the midrange.
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,262
Likes
2,412
Location
Pacific Northwest
How could one alter the signal in the time domain (compression) and leave the frequency domain unaffected? The spectrum must change too.
EDIT: ah, missed the "constant amount". That's sort of a misleading, special case isn't it?
Compression is not constant.
It's a simplifying assumption to illustrate that amplifying the signal, in and of itself, does not change the frequency response, though it does change our perception of frequency response.
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,262
Likes
2,412
Location
Pacific Northwest
The FFT will change only if they EQed it. The FFT will not change if they simply boost the level of quiet parts and do nothing else. You can test this yourself. Take the FFT of a quiet section of music. Boost it 12 dB and take the FFT again. The FFTs will be identical. By this I mean identical in shape / frequency content. Of course the amplitudes of the boosted one will be higher. But they'll be higher equally across all frequencies so the shape won't change.

However, even though the frequency response is the same, we won't perceive it that way. The boosted one, being louder, will sound like it has been EQed to boost bass & treble relative to the midrange.
Correction: boosting the quiet sections will indeed change the overall FFT. It won't change the shape of the FFT of the quiet sections that were boosted. But equally boosting all frequencies in that section, changes the weighted balance of frequencies in the overall waveform (the frequencies in the boosted quiet sections get bigger coefficients in the frequency domain). So the FFT of the entire waveform will show an aggregate change. So differences like those shown above could arise from dynamic compression alone, without any EQ.

That said, even if the overall frequency response doesn't change, we would still perceive it to be different due to the differences in loudness.
 

dc655321

Major Contributor
Joined
Mar 4, 2018
Messages
1,091
Likes
1,391
It's a simplifying assumption to illustrate that amplifying the signal, in and of itself, does not change the frequency response, though it does change our perception of frequency response.

Sure, fair enough.
But a constant shift of amplitudes is not what compression is in the sense used here.
Both the spectrum and one's perception of it would change.
 

Thomas Lund

Member
Technical Expert
Industry Insider
Joined
May 15, 2018
Messages
40
Likes
159
Can a DR compressed album (remastered) sound better (be more faithful) than the previously DR uncompressed version of it?

Yes, but in general, no. It's the original album the artist signed off, and new versions are typically released to make more profit from the back catalogue. Exceptions include Brothers In Arms, where the 1996 version, according to excellent mastering engineer Bob Ludwig, was done without additional old Sony DACs. Peak-to-Loudness ratio (PLR) is slightly lower on the new release, but that might be down to just a generation of conversion brickwall filters being omitted. Anyway, the 1996 version to me sounds better. For instance, listen to the rim shot on the title track.

New versions of classic (pop) albums might be influenced by trends in micro-dynamics (PLR) or macro-dynamics (Loudness Range). If so, we probably have not reached the bottom yet. Find attached a graph of average PLR in the >7000 most popular music tracks from 1963 through 2020, based on Rudi Ortner's thesis and our extension, presented at the AES150 convention, 25 May, 2021.
 

Attachments

  • Deceptive Tracks AES150.jpg
    Deceptive Tracks AES150.jpg
    200.4 KB · Views: 108
Top Bottom