Do I understand “Dynamics”?

Analogous

Well-Known Member
Joined
Nov 17, 2017
Messages
3,343
Points
113
Location
Bangalore
I thought I did and then I read this account.
According to this Dynamics is a basic concept in music that affects everything from writing songs/scores, to mastering and reproduction in an audio HiFi set up..
So…The word is used so often that it can mean different things in different situations.


But in the context of Home HiFi setups. Specifically. How do I look for or perceive dynamic range and decide if it’s reproduced well or not?
 
Only way to know is have a good reference source. The best reference source is original studio recording which most of us will never have access to, the second best and often more accessible source is the early pressings of the song; these could be vinyl or digital depending on how the song was mastered in each media. If one listens to such content on a good enough system one will get a fair idea about how lowest and loudest notes are being reproduced. If one listens to such tracks often enough one will develop a fair idea about dynamic range of a particular recording. A flat or compressed CD/vinyl will sound flat in any system , cost of gear notwithstanding. A good source OTOH will sound pleasing and enjoyable on even an entry level stereo rig, obviously speaker FR will be a limiting factor at entry level.
 
Only way to know is have a good reference source. The best reference source is original studio recording which most of us will never have access to, the second best and often more accessible source is the early pressings of the song; these could be vinyl or digital depending on how the song was mastered in each media. If one listens to such content on a good enough system one will get a fair idea about how lowest and loudest notes are being reproduced. If one listens to such tracks often enough one will develop a fair idea about dynamic range of a particular recording. A flat or compressed CD/vinyl will sound flat in any system , cost of gear notwithstanding. A good source OTOH will sound pleasing and enjoyable on even an entry level stereo rig, obviously speaker FR will be a limiting factor at entry level.
Thanks. That’makes sense.
 
Only way to know is have a good reference source. The best reference source is original studio recording which most of us will never have access to, the second best and often more accessible source is the early pressings of the song; these could be vinyl or digital depending on how the song was mastered in each media. If one listens to such content on a good enough system one will get a fair idea about how lowest and loudest notes are being reproduced. If one listens to such tracks often enough one will develop a fair idea about dynamic range of a particular recording. A flat or compressed CD/vinyl will sound flat in any system , cost of gear notwithstanding. A good source OTOH will sound pleasing and enjoyable on even an entry level stereo rig, obviously speaker FR will be a limiting factor at entry level.
Just having a good source or reference vinyl may not be a sufficient pointer of dynamics in a sound. The speakers should also be capable of reproducing it properly. For ex - speakers not able to reproduce kick drums properly.
In many Jagjit Singh songs particularly those recorded from live shows you will notice that in many systems due to crowd noise, voice of the singer doesnot shine through the system.
Also with classical songs the singer going in from low node to a very high note at a stretch, for many speakers it becomes difficult to reproduce it properly or speakers fail to reproduce it without straining itself.
You need to hear the same song in many systems before you can figure out that a perticular system is missing in dynamics.
 
^ I think you did not read my post, and missed the context, I have clearly said "obviously speaker FR will be a limiting factor at entry level."
Speaker FR also takes room and placement into account. Speaker not been able to produce kick drums properly may also be a function of room and placement. By classical songs, I was pointing towards the content of the music. So whether a system is dynamic or not depends on the Synergy of the system. Here we cannot just point out to source or amplifier or cables.
 
This discussion is an example of the challenges we all face when trying to explain complicated concepts that are applicable in multiple situations and contexts. With several variables impacting the outcome, over simplification is never going to be true or satisfactory.
 
Interesting sort of topical overview and true-enough IMO for what it covers.

The easiest way I ever found to communicate about it to anyone is to ask if there have been times in their life when, walking up to a venue, they can tell whether a band is playing live or not--from the parking lot. Most people have had such an experience.

The one means of demonstration that slaps people in the face is to hear a pair of large 100+ dB/W speakers that couple to a large amount of air. In particular, how they can come alive at low levels. Yes, many things matter, but they pale compared to loudspeaker sensitivity. If you dial them up and go into adjacent rooms or even outside, it's much closer to the preceding experience. Beyond that "big swing", you have to go backwards up the chain. FWIW/YMMV.
 
But in the context of Home HiFi setups. Specifically. How do I look for or perceive dynamic range and decide if it’s reproduced well or not?
I use dynamic range meter ext. (not available from official website)in Foobar 2000 , any pop songs that have a good dynamic range usually falls in the region of 12- 15 dB, above 15dB DR is exceptionally good, espeacially for classical or jazz. Anything else in the single digit DR is too loud , best to be avoided. Songs recorded post 1995 are usually louder with a single digit DR.
List of albums along with their DR

1742532497678.png

 
Last edited:
I use dynamic range meter ext. (not available from official website)in Foobar 2000 , any pop songs that have a good dynamic range usually falls in the region of 12- 15 dB, above 15dB DR is exceptionally good, espeacially for classical or jazz. Anything else in the single digit DR is too loud , best to be avoided. Songs recorded post 1995 are usually louder with a single digit DR.
List of albums along with their DR

View attachment 89958

This is interesting. Would you know what parameters are measured and how this app calculates the DR?
Thanks for sharing this video provocatively titled “why is modern music so bad”😄 The presenter explains all the possibilities lucidly. I could not however understand the “LS” measure he refers to several times. Is it a loudness, clipping and compression metric?
Given that the widely used definition of dynamic range is the difference between the loudest and quietest parts of a piece of music or song, I can understand why the two clips of music he exampled sound so bad.
Or maybe I m just old and don’t get it anymore 😟
 
Last edited:
Would you know what parameters are measured and how this app calculates the DR?
I don't know about the algo calculating DR.
Peak - If its 0dB , the recording is probably loud and "brick walled". There should be some headroom to manage peaks , say -1dB or more.
RMS - even if the recording is of the same track/album ( like different pressings/ releases) has a similar DR, higher RMS means its more louder
LUFS ( You might be refrering to it as LS) stands for Loudness Units Full Scale, a standardized measurement of audio loudness that factors in human perception and electrical signal intensity. Even though theorotically, a 16bit/44K CD can offer 96dB or more DR, vInyl may sounds more dynamic than any modern recordings on a CD/ streaming service, because of the physical limitations of the medium
From the DR Loudness War "Hall Of Shame"-
1742541175518.png
I'm not a studio engineer , some of the above definitions are just my interpretations.
If you want this DR ext. , PM me your gmail id please.
 
Last edited:
I don't know about the algo calculating DR.
Peak - If its 0dB , the recording is probably loud and "brick walled". There should be some headroom to manage peaks , say -1dB or more.
RMS - even if the recording is of the same track/album ( like different pressings/ releases) has a similar DR, higher RMS means its more louder
LUFS ( You might be refrering to it as LS) stands for Loudness Units Full Scale, a standardized measurement of audio loudness that factors in human perception and electrical signal intensity. Even though theorotically, a 16bit/44K CD can offer 96dB or more DR, vInyl may sounds more dynamic than any modern recordings on a CD/ streaming service, because of the physical limitations of the medium
From the DR Loudness War "Hall Of Shame"-
View attachment 89960
I'm not a studio engineer , some of the above definitions are just my interpretations.
If you want this DR ext. , PM me your gmail id please.
Thank you so much. Lots to ponder and try to make sense of here.
 
Thanks for sharing this video provocatively titled “why is modern music so bad”😄 The presenter explains all the possibilities lucidly. I could not however understand the “LS” measure he refers to several times. Is it a loudness, clipping and compression metric?
I won't agree with the fact that modern music is bad.
Even now we get recordings which are stellar and simply touch the emotional chord. But yes most of modern music has become very complex to reproduce and you need a competent system to reproduce it.
For me to reproduce modern day music I need active speakers, tube pre-amplifier, DSP processing to tune the music as per my liking. I have moved on from 15 year old Tannoy Westminster with tube amplifier and turntables (which my father still enjoys) to speakers from professional audio (JBL PRX). I find modern day music through JBLs to be more appealing than the same music played through Tannoys. Tannoy simply fails to bring out the intrinsics of modern music. But then its me. I simply didn't have funds to put in a ultra highend system so took this route.
 
I use ebumeter https://kokkinizita.linuxaudio.org/linuxaudio/ebumeter-doc/quickguide.html

If you are on debian, fedora, etc you can install ebumeter. To install I just did the following on my fedora
$ sudo dnf -y install ebumeter

On ubuntu/debian you will have to do apt-get install ebumeter probably

1. ebumeter is a plugin for jack
2. ebur128 is the command line linux dr meter. e.g.

$ ebur128 --lufs --full 01\ -\ Chopin.flac
ebur128 --full 01\ -\ Chopin.flac
Integrated loudness: 8.0 LU
Loudness range: 10.6 LU
Peak level -0.2 dB
Integrated threshold: -2.4 LU
Range threshold: -12.4 LU
Range min: 1.3 LU
Range max: 11.9 LU
Momentary max: 16.4 LU
Short term max: 15.1 LU

In addition to ebumeter there is another linux tool dr14_tmeter from https://github.com/simon-r/dr14_t.meter. It works for python2 and python3. This is the code that the foobar plugin for drmeter on windows uses. You can install it like this

$ git clone [email protected]:simon-r/dr14_t.meter.git
$ cd dr14_t.meter
$ sudo python ./setup.py install --root="/" --prefix="usr" --optimize=1

$ dr14_tmeter -f "01 - Chopin.flac"
/var/lib/mpd/music/01 - Chopin.flac
/var/lib/mpd/music/01 - Chopin.flac: DR 12
/var/lib/mpd/music/01 - Chopin.flac :
DR = 12
Peak dB = -0.16
Rms dB = -15.85

Using the find linux command you can generate dir14.txt in each directory where music is present. I have modified cantata player to display the DR value when I play a song. The text file dr14.txt looks like this
1742561073730.png
 
Last edited:
I use ebumeter https://kokkinizita.linuxaudio.org/linuxaudio/ebumeter-doc/quickguide.html

If you are on debian, fedora, etc you can install ebumeter. To install I just did the following on my fedora
$ sudo dnf -y install ebumeter

On ubuntu/debian you will have to do apt-get install ebumeter probably

1. ebumeter is a plugin for jack
2. ebur128 is the command line linux dr meter. e.g.



In addition to ebumeter there is another linux tool dr14_tmeter from https://github.com/simon-r/dr14_t.meter. It works for python2 and python3. This is the code that the foobar plugin for drmeter on windows uses. You can install it like this

$ git clone [email protected]:simon-r/dr14_t.meter.git
$ cd dr14_t.meter
$ sudo python ./setup.py install --root="/" --prefix="usr" --optimize=1

$ dr14_tmeter -f "01 - Chopin.flac"
/var/lib/mpd/music/01 - Chopin.flac
/var/lib/mpd/music/01 - Chopin.flac: DR 12
/var/lib/mpd/music/01 - Chopin.flac :
DR = 12
Peak dB = -0.16
Rms dB = -15.85

Using the find linux command you can generate dir14.txt in each directory where music is present. I have modified cantata player to display the DR value when I play a song. The text file dr14.txt looks like this
View attachment 89961
All this is passing right over my head. Any possibility of dumbing it down a bit?

Even better if there are ways to evaluate DR in stereo set ups subjectively?
 
All this is passing right over my head. Any possibility of dumbing it down a bit?

Even better if there are ways to evaluate DR in stereo set ups subjectively?
There is something known as Dynamic Range. It tells you the lowest level in your music and the highest level in your music sample. A good recording will have a wide difference between the lowest level and the hight leve. E.g. You are in a room where people have guitars, drums and they are singing. There is a nail lying on the table and it falls down. When no one is playing anything and the nail falls down, can you hear it?
If the room is noisy you will not be able to hear it. So to make you hear it, let's say I artifically amplify the sound of the nail hitting the floor. Now you will be able to hear it. But such a recording will sound artificial. You brain will immediately recognize it and your enjoyment will drastically reduce.

Now you are driving in a car and listening to a radio station. If the radio station is playing the recording faithfully, you will not be able to hear part of the broadcast where the volume is very low because of the car noise, engine noise, tyre noise, etc. So what the radio station does is that it artificially boosts passage in music with low volume and artificially turns down the volume of passages which are loud. This allows the guy in the car to hear all parts of the music. But it will sound artificial. The volume of vocals will be as loud as the drums and so on. This method of boosting low volume and reducing passages with high volume is dynamic compression. A tool can analyse and find the difference between sample of music with lowest volume and sample of music with highest volume. A good recording will show a good difference, but you will be able to hear and appreciate this music only with a good quiet room in a village with a good system. A compressed recording will show very little difference, but you will be able to hear this music in a noisy atmosphere and any noisy city like Bombay, Bangalore, Delhi.

Audiophiles are so fickle minded that they worry about brand of the equipment, rca connectors, cable risers, cables, mp3, flac, hi-res, bit perfect. But they forget that an even an MP3 with high dynamic range will beat a flac with low Dynamic range.

I can dumb down if you have one of the following

1. A macos box. Check this https://www.audiosciencereview.com/...alyzing-dynamic-range-of-music-library.35935/.
One important tool mentioned is DR Offline MKII by MAAT.

2. A Linux box. Even a raspberry PI will do. You can simply install ebumeter. Once you install ebumeter, it will install an executable ebur128. You can use ebur128 on your music files to tell you the DR range in the music file.

This is valiid if you primarily listen to your local music collection instead of spotify, apple music, qobuz, tidal FLAC. If you are a streaming person you can't use any of the tools and you are at the mercy of the streaming provider.

You can go through these posts. It will clear lot of your doubts




 
Last edited:
All this is passing right over my head. Any possibility of dumbing it down a bit?

Even better if there are ways to evaluate DR in stereo set ups subjectively?
Here is some information on what streaming service providers do. Bigger the number (disregard the negative sign), better it is. For recordings that have very good Dynamic Range, Spotify Premium (if you choose the quiet setting), will sound better than Apple Music regardless of Hi-Res and what not. Apple Music uses fixed Normalization of -16 LUFS. Spotify premium allows you to choose -11 dB LUFS for loud. This will be worst quality but good for listening in a noisy environment. You can chose -19dB LUFS which will give you better dynamic range and the same song may sound better an apple music if the original master has good dynamic range in the recording. But to take advantage of Spotify's -19dB LUFS, the master submitted to Spotify has to be good. To quote "Darko"

Much hoarse shit gets shovelled into Gucchi bags about high-resolution recordings which “finally offer playback at real-life dynamic range”. In theory, the argument is rosy indeed. In practice, it’s sharp thorns all the way when heavy dynamic compression during particularly the commercial mastering process squashes dynamic range to death.


Table taken from https://en.wikipedia.org/wiki/Loudness_war

1742613326557.png

How Spotify adjusts loudness

1742613498798.png
 
Last edited:

You can enter the name of an album and check its dynamic range for the various cd and vinyl releases. The dynamic range for most popular albums will be available on this site.

For example Adele 21, the cd has an average dynamic range of 6 whereas its vinyl release has an average of 11. So in this case the vinyl will definitely sound better.

Basically you’ll know which press to buy and whether to get the vinyl or the cd.

Normally jazz and classical music will not suffer from dynamic compression.
 
Last edited:
Here is some information on what streaming service providers do. Bigger the number (disregard the negative sign), better it is. For recordings that have very good Dynamic Range, Spotify Premium (if you choose the quiet setting), will sound better than Apple Music regardless of Hi-Res and what not. Apple Music uses fixed Normalization of -16 LUFS. Spotify premium allows you to choose -11 dB LUFS for loud. This will be worst quality but good for listening in a noisy environment. You can chose -19dB LUFS which will give you better dynamic range and the same song may sound better an apple music if the original master has good dynamic range in the recording. But to take advantage of Spotify's -19dB LUFS, the master submitted to Spotify has to be good. To quote "Darko"




Table taken from https://en.wikipedia.org/wiki/Loudness_war

View attachment 89972

How Spotify adjusts loudness

View attachment 89973

Thanks, I just tried it out and the 'Quiet' setting certainly has less 'hash' on complex passages.
 
Check out our special offers on Stereo Package & Bundles for all budget types.
Back
Top