• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Inside High-res audio: PCM vs MQA vs CD: 2L Sampler Comparison

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,445
Likes
15,781
Location
Oxfordshire
And you thought his comment that no one has heard -60 db not???
I don't agree with it but suspect it may be a translation problem from somebody who is not a fluent English speaker.

Edit: It seems it is a question of greatest actual dynamic range @Frgirard has seen on a recording and IME that won't be far out. I have heard of a recording with 70dB but only one.
 
Last edited:

Frgirard

Major Contributor
Joined
Apr 2, 2021
Messages
1,737
Likes
1,040
I don't agree with it but suspect it may be a translation problem from somebody who is not a fluent English speaker.

All my plugins show between - 48 and - 72 dB.
The avg is - 60. A professional deformation.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
I don't agree with it but suspect it may be a translation problem from somebody who is not a fluent English speaker.

Edit: It seems it is a question of greatest actual dynamic range @Frgirard has seen on a recording and IME that won't be far out. I have heard of a recording with 70dB but only one.
I've run across a couple in the mid 70's. 74 db for one and I think 76 db for the other. That is using one number of course and realizing how that can lead one astray about audible noise levels.
 

Grooved

Addicted to Fun and Learning
Joined
Feb 26, 2021
Messages
679
Likes
441
I don't agree with it but suspect it may be a translation problem from somebody who is not a fluent English speaker.

Edit: It seems it is a question of greatest actual dynamic range @Frgirard has seen on a recording and IME that won't be far out. I have heard of a recording with 70dB but only one.

Just checked one of the few files I tested these last days : Thomas Strønen - Bayou - FLAC 24-96 Qobuz
I even cut the lowest parts at the beginning and the end and redo the analysis : you still have more than 60dB between lowest and highest RMS and 5 more to peaks
Analysis.PNG
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,375
Likes
234,488
Location
Seattle Area
All my plugins show between - 48 and - 72 dB.
The avg is - 60. A professional deformation.
Your plugins are wrong. There is no way to properly determine dynamic range of music as its level can sink below noise floor. The tools you use time window averaging which doesn't work for this purpose. Here is Adobe Audition Amplitude Statistics Settings for example:

1621813030307.png


50 miliseconds of averaging would easily burry low level samples. Here is a random file in my library being analyzed with that setting:

1621813131613.png


Now see what happens if I changed that to 1 millisecond:

1621813182700.png


See how it now shows 10+ dB lower sample values.

Unfortunately it won't go any lower than 1 msec so you can't see the true lower floor. With that setting, it still shows much better dynamic range than you claimed:

1621813305677.png


Without averaging, any silence in the clip would set the lowest floor, giving you huge dynamic range. This could be faulty of course because it could be fade to silence. Alternatively noise/dither could hide true correlated musical content.

Net, net, don't run off with tools without understanding what they really do. Just because a program spits out a number it doesn't mean it is what you want.

The only solution here is proper analysis of recording venue/room noise and max loudness which is referenced in the papers I showed you.
 

Attachments

  • 1621813264735.png
    1621813264735.png
    15.7 KB · Views: 65
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,375
Likes
234,488
Location
Seattle Area
I don't agree with it but suspect it may be a translation problem from somebody who is not a fluent English speaker.

Edit: It seems it is a question of greatest actual dynamic range @Frgirard has seen on a recording and IME that won't be far out. I have heard of a recording with 70dB but only one.
Neither one of you has the tools to make this analysis. None exists that provide the values. The only person I know that has developed such a tool is Bob Stuart of Meridian MQA fame! It came up in my discussions with him some 20 years ago. More recently I read a white paper by the people who built musicscope analyzer trying to address MQA. They did not disclose the data but said to have developed a tool for such statistic analysis.

Bottom line, he is just repeating all the myths out there about this topic. I wrote that article for WSR magazine because I got tired of writing the same corrections and quoting the authoritative papers on the topic. If you have a paper that shows such low dynamic range in music with proper analysis, let's see it. Otherwise, let's follow the science and not spready common objectivists myths that are just as wrong as some subjectivist ones.
 

Frgirard

Major Contributor
Joined
Apr 2, 2021
Messages
1,737
Likes
1,040
The real life is an issue in the audiophile world.

Bob is God so ite missa est.
 

MWC

Active Member
Joined
Apr 15, 2019
Messages
153
Likes
64
https://www.sonicvisualiser.org/download.html does not accept DXD either. It also seems to down-convert 192kHz PCM to 44.1kHz at least for playback. Although it might well be pilot error, as I just now downloaded this program and have yet to take my driving test in it,
 
Last edited:

krabapple

Major Contributor
Forum Donor
Joined
Apr 15, 2016
Messages
3,169
Likes
3,717
Neither one of you has the tools to make this analysis. None exists that provide the values. The only person I know that has developed such a tool is Bob Stuart of Meridian MQA fame! It came up in my discussions with him some 20 years ago. More recently I read a white paper by the people who built musicscope analyzer trying to address MQA. They did not disclose the data but said to have developed a tool for such statistic analysis.

Bottom line, he is just repeating all the myths out there about this topic. I wrote that article for WSR magazine because I got tired of writing the same corrections and quoting the authoritative papers on the topic. If you have a paper that shows such low dynamic range in music with proper analysis, let's see it. Otherwise, let's follow the science and not spready common objectivists myths that are just as wrong as some subjectivist ones.

WSR -= Widescreen Review?

I'd like to read your article. Can you (re?)post a link?
 

Phos

Member
Joined
May 27, 2021
Messages
60
Likes
39
I was messing around to see if I could get Audacity to do something similar, it turns out that The Doors probably wasn't a good candidate for high res audio:
1623488328363.png

This was recorded in the 60's, so I have to imagine that tone is the capstan or slack take up motor (seems to only be on the guitar track, hey-o early use of stereo mixing). It is down at -68 dB. The symbol hits do actually reach into the higher frequencies, as you would expect. I got this version from HD tracks because the version I already had (itunes maybe?) had a bunch of crackling in it for some reason that I only caught once I upgraded from my ATH-A900's.

Edit: Wow this Bob Marley album has remarkably good mastering from what I can see:
1623491907694.png

And since I happen to also have Rain by The Lake Poets from B&W's service,
1623492235719.png

And yeah there's that peak.
 
Last edited:
Top Bottom