Dynamic range true value

[Doctor Doubledrop]

Hi,

any expert engineer that well know how to misure the dynamic range of a track?
I have Maat DR meter and Brainworx bx meter, they shows not small different values of the dynamic.. it results impossible to me monitor it in this way.

Maat support replied me this:
two different measurement methods. There is the PMF DR measurement standard, first seen in the TT DR Dynamic Range meter and TT DR Offline Meter and used with permission in MAAT’s DR line including DRMeter, DRMeter MkII, DROffline and the soon to be released DROffline MkII. The unauthorized Brainworx meter which, as you noticed, doesn’t measure DR and conforms to no known standard.
Accurate (and official) DR is measured only with a PMF or MAAT product; simple as that!

plugin alliance support replied me this:
I heard back from the dev guys and they basically said that the meters simply have different integration times and there is no golden standard when it comes to measuring the dynamic range of an audio signal. It seems like our meter has a slightly faster time that it uses for its calculation. So it is expected behavior and you can use both readings with caution as they are somewhat relative and not absolute values like the Peak and RMS values on dBFS scale.

Now.. is not possible to have different values, for sure 2 different tools could have 2 different methods.. but it is not possible that a car with two different odometer shows 80 km/h for one and 110 km/h for the other.. am I wrong?

Here's few screens about the test:





Regards

[MogwaiBoy]

That sausage looks hella squashed so why care about DR?

Realtime dynamic range analysis is a mixed bag. You could try the 'Crest' reading in Voxengo SPAN or the app EXPOSE by Mastering the Mix does neat stuff offline (the TT Dynamic Range and MAAT meters also have offline versions). I think offline will give you a more accurate reading.

[Doctor Doubledrop]

The master in the screens is just an example of course, and even if it looks like a sausage u can easy see really not small difference between the two meters.. I don't have Mastering the mix but in SPAN.. ah I've just learned another thing ahah.. I usually use it for spectrum analysis hiding meter and stats.

Anyway I always check DR not only for the full mix as in the example, I always monitor it when I use compressor, tube or saturator.. but this is not the point of the thread.. the point is that there could not be two difference values.. as said is not possible that a car with two different odometer shows 80 km/h for one and 110 km/h for the other.. what is the real value? they are no free tools, so I'd like to pay for something for what I can trust and for a tool that do its work well.. at the moment basically I've thrown my money into the trash..

[Ah_Dziz]

There are different measurements of dynamic range. You could say a car has its engine running at 1000 rpm and the same engine running 60000 rph. Both are correct they are just using a different scale which leads to a different measurement. If by dynamic range you mean the difference between the loudest a song ever gets and the quietest it ever gets then that should be easy to measure. Coming up with a meaningful realtime measurement that gives you that sort of info can be done in many ways and will not give you the same results. Even using an rms measurement of the level of the track can give different results as one changes the rms time over which the level is averaged.

[Googly Smythe]

I think you would be better off with a lufs meter - at least it would use internationally recognised standards. The Youlean one is generally well regarded, and is free.

Accurate (and official) DR is measured only with a PMF or MAAT product; simple as that!

Well, they would say that, wouldn't they.

[Doctor Doubledrop]

Honestly I continue to don't understand.. there are several compressor that shows gain reduction, each one with its own method, but value is the same for all.. there are several spectrum analyzer, the spectrum looks always the same even if with different graphic visualization.. and go on.. I really don't understand.

About the lufs meter I think that I should leave it for my mastering engineer, it has no benefit for me during the mixing process.. or I may am wrong?

Well, they would say that, wouldn't they.

this help me to don't trust in Maat reply, even if doesn't help me to use one or the other for monitor the DR.

[plexuss]

I'll try again to address this question...

First, this is just what I have come to understand, not the absolute truth. I think I am accurate within a workable margin of error. But definitely correct me anywhere I am wrong (and provide citations if you can).

The TL;DR of Dynamic Range

DR is measurement, usually, of some peak value vs some averaged value. What is measured specifically is based on the people who implement the meter. There are standards out there for various metering techniques, so if those are employed you end up with a better sense of what exactly is being metered. Here's an example...

Loudness vs peak aka perceptual vs technical metering. Perceptual metrics are based on how the audio will be perceived. Technical metrics are based on the electrical nature of the signal. For example "peak" is usually the absolute highest or lowest value the signal reaches in terms of is measured value, usually in volts, converted to digital and then scaled with reference to the loudest possible sound that can be digitally created. That sound is where all bits are ON or 1. 16 bit audio, 16 bits are on. 24 bit audio, 24 bits are on. You can't go higher than that (there are some outlier exceptions to this). This is 0dBFS (full scale). This is an absolute measurement. And from this, all other levels are referenced as a ratio of the full scale, relative to 0dB.

[Side note: since audio is only relevant when it's tranceived through a medium, like air, it needs to be in an analogue format. so ultimately all aspects of audio can be measured in the analogue domain. well, all the important things. so when you are measuring digital, there is an analogue counter-part that is ultimately more important. this is why loudness measurements came to be... i'll get to that...]

The measurement made relative to 0dBFS is a "technical" metric because it relates to the physical nature of the signal (not necessarily how we perceive it). This is because our hearing is not technically neutral (flat, linear response, lack of distortion, etc). Google Fletcher Munson. We have to mathematically transform the much easier to measure technical metrics into numbers that represent how we hear.

!!!

This is a really amazing thing when you think about it... so take a moment to think about it.

This is what loudness metering is: a numerical representation of audio that is relative to how we hear. Both sets of metrics are important: technical metrics show the audio behaving relative to a physical absolute (eg. flat response, 0% distortion, etc). Perceptual metrics show how the audio is behaving relative to how humans will perceive the audio.

The perception of the audio is different from the technical make-up of the audio, except through a mathematical transform that relates them together. These are usually in the form of industry standards from the like of EBU, IEEE, etc. But anyone is free to create a transform - there are many independent ones out there. Of course there are benefits to using standards...

As this relates to dynamic range (DR): There will be two ways to measure DR: technically and perceptually. What is the DR metric? Off the top of my head, it's just that! a measurement from the peak of something to some other metric of the signal. That's a general definition. More specifically it could be to the peak of the audio within a given time-frame and the average of that audio within the same time-frame. This is essentially what it is. The ratio of the peak to the RMS average within a given time-frame. With RMS based DR there is a standard that defines the exact way this is to be done and what the time-frame is. This is what you get with technical DR meters that aren't using a loudness standard, which is the next topic. Loudness DR is the same thing but relative to a scale of loudness metering. A loudness meter is all about how the audio will sound to a person. And so, the loudness based DR is about how the dynamic range will sound to a person. Compared to the RMS DR which is the dynamic range in a technical numerical physics science sense. Both are useful!

There are many standards for each group of technical and perceptual metering. And there are many other ways of doing metering beyond those standards. For example Dorrough use their own metering algorithm that is considered a good contender - my perception of Dorrough is that it tries to incorporate both some sense of technical and of perceptual into one meter. But you sacrifice some accuracy perhaps...

Anyway, that's all I got. Hope I am not too wrong about stuff and this was helpful.

Some docs:
EBU R128 Loudness normalisation and permitted maximum level of audio signals PDF

ITU-R BS.1770-4 Algorithms to measure audio programme loudness and true-peak audio level PDF

[Doctor Doubledrop]

hey mr. plexuss! what a reply

basically u have not helped me with the thread question .. but u have well explained to me something that I'll would have look in something like 10-20 years I think ahah

interesting the Fletcher Munson curve too, never heard about it.. and I was happy to realize that even if I didn't know it I always mix in the right way.. I think that before or after a musician will realize it.. unfortunately this help u yes, but if u don't know how to eq the sound u will continue to make music that will be improved time by time and not perfect mix.. but first if we will learn all immediately where is the goodness? second I don't want to go out of the thread purpose.. but as said I've really appreciated your reply and the time u took to write it!


Many greetings man! the best with your production!

[Doctor Doubledrop]

Just a quick update from Maat

DR (Dynamic Range) was introduced and initially promoted by the PMF. Friedemann is the founder, he found that there was no standardized method for gauging “dynamic density,” or the amount of aggressive dynamic range reduction being applied to popular music. So, he worked with some colleagues and created the DR standard. MAAT is granted exclusive license to continue DR tool development.

how can I don't trust in them?

[plexuss]

Just a quick update from Maat

DR (Dynamic Range) was introduced and initially promoted by the PMF. Friedemann is the founder, he found that there was no standardized method for gauging “dynamic density,” or the amount of aggressive dynamic range reduction being applied to popular music. So, he worked with some colleagues and created the DR standard. MAAT is granted exclusive license to continue DR tool development.

how can I don't trust in them?

It's not a matter of trust. I tried to explain why a dynamic range measurement needs to be made understanding the specific algorhythm used. There is no one true way to measure DR. You may have to do more learning of the foundational stuff before you can understand what is it your measuring and why.

[MogwaiBoy]

Why? Well there is this...

http://dr.loudness-war.info/

Which incidentally states "MAAT DROffline log file format is now supported".

Which incidentally MAAT just released a new "MkII" version of, shafting V1 owners with no upgrade path.

[plexuss]

You don't HAVE to use an expensive meter to measure DR. For example right now Youlean's meter is free and offers loudness based DR measurement (PLR/PSR) which will give you a measurement based on audio indsustry standards for perceieved dynamic range. You also have bx_meter which is a more technical measurement. And you have MAAT which is based on R128 etc. So you already have what you need: bx_meter for a technical DR measurement and MAAT for perceptual DR measurement based on industry standards. The fact that MAAT states this means you can trust them.

[Doctor Doubledrop]

Hi plexuss, I think I've understand what u mean.. but I still can't understand.. sorry ahah!

I always hear "that track has a good DR, 7", "that track has just DR 3 but sounds good at all" and so on.. so my obvious question is "ok, but what are u talking about?, what did u use for read the DR?" maybe as said the offline Maat version is the answer.. but is it really the standard used by all when they talk about the DR of their music?

anyway the br_meter shows me always smaller values of DR and see 3 or 4 in the mix sounds so strange to mee.. what will be the master DR?? 1??

anyway, as said.. yes I understand what u mean.. but there should really be a confirmed standard.. and Maat says that there's a standard and its their one.. I'm so confused!!

[marzelli]

anyway, as said.. yes I understand what u mean.. but there should really be a confirmed standard.. and Maat says that there's a standard and its their one.. I'm so confused!!

You mean there's a company claiming they've invented measuring DR and their methods are the right ones? Haha seems totally legit I wonder how many have falling for such a marketing trick.

[Doctor Doubledrop]

You mean there's a company claiming they've invented measuring DR and their methods are the right ones? Haha seems totally legit I wonder how many have falling for such a marketing trick.

perfect…
is there any Alan Parsons or similar that tomorrow can proclaim a definitely standard for all? so when someone say that the Dr of that track is "x" anyone knows what he is talking about??