• 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.

OP
S

skymusic20

Member
Joined
Aug 19, 2020
Messages
40
Likes
15
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.

This is sad. Imagine you are a young person (Im certainly not one) and you want to listen to the best recordings of 50's, 60's, 70's and 80's music.

All you find online (or physical stores) are remasters and more remasters...

Sure you can go to the used market (online or physical) but it is wild and very time consuming...
 
OP
S

skymusic20

Member
Joined
Aug 19, 2020
Messages
40
Likes
15
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.

Thanks for the graph... Very sad but interesting...

By the way, according to your graph, my remasters of Scorpions "In Trance" and "Taken by Force" albums are pure junk...
 

Thomas Lund

Member
Technical Expert
Industry Insider
Joined
May 15, 2018
Messages
40
Likes
155
If you want low compression then... good luck!

Luck is good, but it's possible to measure your tracks as well. However, allow me to start from the pedantic side. We should try to distinguish between content and the reproduction system. "Dynamic range" could be reserved for the latter: A ratio expressing the highest level (magnitude) vs the lowest level a system conveys.

To reduce ambiguity, use a different terminology about the content, like we say "height" about an object and "clearance" about where the object might fit. Peak-to-Loudness Ratio (PLR) is a defined way of expressing micro-dynamics (squashing) in music, and it also relates to peak level vs average level, rather than peak vs noise floor.

To measure your own tracks, visit http://musictester.net/demo and drag files (WAV, AIFF, AAC, MP3 etc.) into the window. One or more tracks may be compared (and listened to) side by side, see attached. It's free to use, and the four columns measure the parameters specified. "PLS" means Peak-to-Loudness Short-term, based on the most squashed 10 secs of the track (black line in the histogram). By switching "Normalization" to "Loudness", you can listen to music at equal Loudness, e.g. two different versions of the same track.
 

Attachments

  • Deceptive Tracks AES150 p11.jpg
    Deceptive Tracks AES150 p11.jpg
    162.1 KB · Views: 35

Frgirard

Addicted to Fun and Learning
Joined
Apr 2, 2021
Messages
935
Likes
537
A certain amount of compression can be very positive to help set the volume level with playing with it during playback as it goes through the roof or drops away.
The commercial music is made to be sold to as many people as possible and to listen anywhere whatever the level of noise in the environment.

The high compression without the good tonal balance (to hide the high level of distortion and to make the sound loud) doesn't mean nothing.
 

cany89

Active Member
Joined
Dec 22, 2020
Messages
238
Likes
115
Just as an example, take a look at two of my fav all time albums:
Scorpions In Trance:
https://dr.loudness-war.info/album/list/year?artist=scorpions&album=in+trance

Damn, I have never seen anything like this before? What did they use? Soundblaster card and basic iPhone headphones?
The average is 5! The lowest is 3!!!???

Screen Shot 2021-08-19 at 13.02.25.png


You are right btw, it's getting more expensive to buy older and even more damaged CDs. I think I'll sell my CD player. I got over 200 CDs. It's probably nothing for some. But for me, they are all special. I'll keep those. I ripped them bit-perfect FLAC anyway. And now listening to Qobuz most of the time.
 

JJB70

Major Contributor
Forum Donor
Joined
Aug 17, 2018
Messages
2,436
Likes
5,207
Location
Singapore
The commercial music is made to be sold to as many people as possible and to listen anywhere whatever the level of noise in the environment.

The high compression without the good tonal balance (to hide the high level of distortion and to make the sound loud) doesn't mean nothing.

I am just commenting based on my own experience as a listener. The dynamic range of an orchestra is quite extraordinary. Indeed the dynamic range of many solo instruments is wider than some imagine. If that dynamic range is fully captured and you adjust volume to give a reasonable level at the bottom end it is going to be extremely loud at the other end. Now that is accurately capturing the range of the performance but whereas I don't find it an issue in a concert hall I find it can result in volume levels I find a bit excessive on audio equipment.
 

Offler

Active Member
Joined
May 27, 2021
Messages
194
Likes
156
From my experience, a lot of music genres do not use "silence" or much of the dynamic range on purpose. Bad score in DR does not automatically mean that the recording is bad. Its just meant to be like that...

Also there are records which indend to "envelope" the listener in sound. That can be best achieved using either headphones or near-field speakers and recording with low DR values.

Then there are all other scenarios:
Classical music, live recordings, acoustic recording etc...

There is usually a lot of silence in the recording and its expected that noise in the surroundings is either supressed (headphones) or the listening room is quite silent. For example an industrial metal band can have DR score of 6-7 on studio album, but live recording of the same song is on 12.

Remasters which do not include re-recording will most likely sound worse compared to the original one.

Also there are recordings which have good DR score, but they sound bad anyway - i have Metallica albums with DR12, but there is simply no bass, unless volume is on maximum and bass-booster/equalizer used...
 

dasdoing

Major Contributor
Joined
May 20, 2020
Messages
1,694
Likes
859
Location
Salvador-Bahia-Brasil
Sounds unbelievable. What track or tracks have that kind of DR?

First classical work I liked was Bolero from Ravel. That was like 20 years ago. I ripped it of a CD from my father. if you adjust the volume at the beginning to medium level, the finale was wayyy too loud. you had to adjust the beginning to barly audible. I actualy used it as an alarm clock. it was perfect to be wakened naturaly
 

dc655321

Addicted to Fun and Learning
Joined
Mar 4, 2018
Messages
985
Likes
1,189
First classical work I liked was Bolero from Ravel. That was like 20 years ago. I ripped it of a CD from my father. if you adjust the volume at the beginning to medium level, the finale was wayyy too loud. you had to adjust the beginning to barly audible

Neat, but I was asking for quantitative evidence of works with a dynamic range of 90dB, as claimed previously. I'm of the opinion that's not even close to realistic...
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,134
Likes
2,248
Location
Pacific Northwest
Sounds unbelievable. What track or tracks have that kind of DR?
Some performances of Tchaikovsky's 1812 fire actual cannons in the crescendo. That should exceed 90 dB from the quiet parts. Of course, it's debatable whether this work actually is "music" ;) but it will be performed and recorded as long as audiophiles exist.

Orff's Carmina Burana also has some serious dynamic range. In my recording of San Francisco & Blomstedt, the quietest parts are -60 dB with peaks a fraction below 0. This is about the biggest dynamic range I've seen with actual musical instruments (not cannons).

Sitting in front of the brass section in community band (why I wear musician's earplugs), I can vouch for live performances of acoustic music hitting 120 dB SPL at least close up. I would never play it back that loud on my stereo, what's the point of blasting it so loud you have to wear earplugs?

However, the music doesn't actually need 90 dB of dynamic range in order for broad spectrum (not noise shaped) dither to become audible as low level hiss. Depending on the frequency spectrum of what you're listening to, that dither may become audible even when the music is somewhat louder. Put differently, you may hear that hiss "down in the mix" in very quiet sections of the music. IME, music at -60 dB is not enough to hear the dither, so the threshold must be somewhere between 60 and 90.

This is why I'd describe 16-bit as musically transparent for most practical purposes, but not perceptually transparent.
 

dasdoing

Major Contributor
Joined
May 20, 2020
Messages
1,694
Likes
859
Location
Salvador-Bahia-Brasil
Neat, but I was asking for quantitative evidence of works with a dynamic range of 90dB, as claimed previously. I'm of the opinion that's not even close to realistic...

it obviously wouldn't make any sense, since -60dBfs is allready unhearable in most listening rooms
 

Frgirard

Addicted to Fun and Learning
Joined
Apr 2, 2021
Messages
935
Likes
537
dBfs is not equal dB spl. If the fancy takes me I can have for -20dbfs, 80 dBspl or 105 dBspl
 

dasdoing

Major Contributor
Joined
May 20, 2020
Messages
1,694
Likes
859
Location
Salvador-Bahia-Brasil

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,134
Likes
2,248
Location
Pacific Northwest
... What I mean is that with reasonable volume you hardly (if even) can hear isolated -60dBfs sounds (let alone masked). test with your listening volume for example here: https://www.audiocheck.net/audiotests_dynamiccheck.php
Sure, 60ish dB is in the ballpark for short-term dynamic range perception, which is what the test you linked does. Yet our hearing has built-in biological dynamic range compression that shifts our perception to match the loudness of ambient sounds. So our long-term dynamic range perception is wider. Think of our loudness perception as a window roughly 60 dB wide that we can shift up or down to match the environment. For example if you go into a very quiet environment like an anechoic chamber or deep underground cave, after several minutes you can hear your own heartbeats.
For musical transparency, 60 dB is likely enough for most music. But for perceptual transparency, whatever format and equipment we use for audio should handle our long-term dynamic range perception, something like 120 dB, not just our short term dynamic range perception.
 

Wes

Major Contributor
Forum Donor
Joined
Dec 5, 2019
Messages
3,851
Likes
3,561
Sure, 60ish dB is in the ballpark for short-term dynamic range perception, which is what the test you linked does. Yet our hearing has built-in biological dynamic range compression that shifts our perception to match the loudness of ambient sounds. So our long-term dynamic range perception is wider. Think of our loudness perception as a window roughly 60 dB wide that we can shift up or down to match the environment. For example if you go into a very quiet environment like an anechoic chamber or deep underground cave, after several minutes you can hear your own heartbeats.
For musical transparency, 60 dB is likely enough for most music. But for perceptual transparency, whatever format and equipment we use for audio should handle our long-term dynamic range perception, something like 120 dB, not just our short term dynamic range perception.

can you be more specific? not clear what you mean there - if you mean a 'mechanical' means in the ear, then yes

normal human DR is on the order of 140 dB, BTW tho that's a f(freq.)
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,134
Likes
2,248
Location
Pacific Northwest
Muscles in the ear shift the positions of the bones to change their sensitivity. That's how it slides the window of loudness perception to adjust for ambient noise. There may be other factors as well, this just one that I know of.

The important point is that tests like the one above vastly underestimate the full dynamic range of our hearing because they only test short-term dynamic range perception. That's important for some things, of course. But it's not the full story.
 

dasdoing

Major Contributor
Joined
May 20, 2020
Messages
1,694
Likes
859
Location
Salvador-Bahia-Brasil
you can hear your own heartbeats
I guess this is a myth lol. never been to a chamber, but it makes no sense to me. What you will hear is ringing caused by ear damage and your breath.


long-term dynamic range perception

don't know if this is realy a thing, but I think you are confusing things. the reason you hear very quiet sounds in very quiet ambients is that they are very quiet ambients. in a nomal living room, even living in a quiet place you will always have a noisefloor
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
2,134
Likes
2,248
Location
Pacific Northwest
... don't know if this is realy a thing, but I think you are confusing things. the reason you hear very quiet sounds in very quiet ambients is that they are very quiet ambients. in a nomal living room, even living in a quiet place you will always have a noisefloor
You refer to simple masking, which does happen. But you're missing my point, which is about psychoacoustics, human perception of loudness/dynamics. Our full perception of dynamics is much wider than the dynamic range we can perceive in any given short window of time.

For example, read section 2.1.b here: http://www.gedlee.com/downloads/HT/Chapter 2.pdf

Tightening those muscles changes the response of the middle ear bones, reducing the amplitude of loud sounds, which extends the total range of dynamics/loudness we can perceive. However, when they're tightened we can't hear the very soft sounds that we can hear when those muscles relax.
 
Last edited:
OP
S

skymusic20

Member
Joined
Aug 19, 2020
Messages
40
Likes
15
Remasters which do not include re-recording will most likely sound worse compared to the original one.

Also there are recordings which have good DR score, but they sound bad anyway - i have Metallica albums with DR12, but there is simply no bass, unless volume is on maximum and bass-booster/equalizer used...


Excuse my ignorance... What is "re-recording" and how can I know if a remaster includes this or not?

By the way, I have Mike Oldfield Five Miles Out with average DR=13 and it really lacks punch and bass. You have to really turn up the volume to enjoy it. There is no bass. I guess "Five Miles Out" song was intended to be punchy but the CD does not convey that...

Thanks
 
Last edited:
Top Bottom