Dynamic Range in Amplitude Statistics

[danrodv]

Hello forum!

Where can I find a more technically detailed explanation of the following amplitude statistics items:

Dynamic Range

Dynamic Range Used

Maximum and minimum RMS amplitude

Also, ALL of the RMS settings options (window width specially)

The information and explanations I find in the "help" really don't tell me much. For example, the "windows" used for the RMS statistics overlap? If so, by how much? Which is the default value for this setting? For which values a window is considered to be "wide" or "narrow"?

I have analyzed audio programs with excessive "treatment" for loudness (dynamic range compression, common in pop music) which show dynamic range statistics comparable to programs which are not processed for loudness (for example pieces of classical music).

I set the window width as recommended by Adobe (wide for wide dynamic range and narrow for narrow dynamic range) and obtained the results found in the attached screen capture.

I find it hard to believe that the pop music program - which is excessively processed for loudness, shows a wider dynamic range than that for the classical piece of music!

I find these results very confusing.

Your help is much appreciated!

[SteveG_AudioMasters_]

I'm afraid that you won't find any more explanations than you already have. People have been asking about this for decades now, and nothing more has been forthcoming. But I have to say that your comparisons don't really mean a thing because we have no idea of what your samples comprised - you only have to include a bit of silence for the apparent dynamic range to shoot up.

[danrodv]

Thank you Steve!

I regret there's no more information regarding dynamic range statistics, but hey! Perhaps I can be of help! What would it take to interact with Adobe to work on this problem? (document further amplitude statistics) I also can help you with the translaations to Spanish  🙂

I see your point. I will proceed then to make some tests taking only portions of the program which do not contain silences or lower program amplitudes and see what happens...

By the way, my statistics comprised the left channel of two complete audio programs - one piece of classical music and a pop song. Please see attached screen captures.

Thanks so much again!

[SteveG_AudioMasters_]

You really won't get anywhere with Adobe! Let me give you a clue as to how difficult this is. Audition was originally Cool Edit, and effetively written by one man - David Johnson. We were in contact with him on a regular basis, and even got some explanations about some parts of the software - but he would never explain the background to the amplitude statistics. Now it's just embedded in the code - I think that there's a pretty good chance that none of the present development team could actually answer questions about it anyway, as they just include that code fragment into new builds. And even if they do know - Adobe would regard any algorithms involved as proprietary, so they wouldn't tell you anyway. I've had a good relationship with the development team for over two decades now, and if they won't tell me - well I'm afraid you stand no chance!

[danrodv]

Oh, well that is rather sad.

I once met Durin Gleaves and had the opportunity to make some suggestions to him. Particularly, to consider the inclusion of crest factor in the statistics. I exchanged a few e-mails with him afterwards, showing how easy this could be achieved. But then followed only silence... I understand now.

As I said, this is rather sad because I think Audition offers some terrific analysis tools, and with further documentation I think you could maybe develop new analysis applications, not to mention that when it comes to education, Audition is a very attractive environment that notably helps in the teaching of digital audio basics to new students.

Thanks Steve!

[SteveG_AudioMasters_]

The other reason that you will get silence from Durin is that he's not an Adobe employee any more. He's left for pastures new...

[danrodv]

Yes Steve, I know. I learnt that from him a couple of weeks ago.