WHAT RMS FOR YOUR TRACKS?

[Angstrom]

the Root Mean Squared (average) db values of an entire track will indicate how much dynamic variance there is in that recording. So a track which has it's quietest moment at -6db and it's peak at 0db might have an RMS of -3db. That's pretty squashed.
meanwhile a track that has a quietest moment at -30db and a peak at 0db might have an RMS of -15

of course, I'm oversimplifying and using a nasty peak value, but that's roughly how an RMS value can indicate the dynamic range of your song.
-3db = nearly no dynamics, -15db rms = plenty of dynamics

disagree.

RMS is simply the Root Mean Squared value of a signal. it doesn't imply decibels or any use of the measurement, nor does it imply anything about dynamic range. for a pure sine wave it means about 2/3 the peak value (sqrt(3)/2) IIRC it's morning here. it's a measurement of effective energy of a signal rather than it's pure peak value.

to get my RMS I load my final song into my wave editing application and look at the statistics for the whole wave. This will give the RMS for the whole song. RMS is an average through time, because obviously you need more than one point to average. So you can see here it tells me where my peak is(3minutes 19 secs), what that peak is (-0.3db) and what the average of the whole song is (-15db):

[Tone Deft]

there is no "The RMS" RMS is an adjective.

I wouldn't use RMS as a peak indicator, it's not the absolute peak level it's some level below that averaged over time. peaks cause clips, so you need to watch for absolute peaks on you meters.

my $0.02. use what works best for you, just correcting some semantics.

[Angstrom]

to talk about "the RMS" of something is perfectly acceptable. Exactly like saying "the average of the results".

the square root of 92
the average speed of the runners
the mean height of the jumps
the RMS of the wave amplitude

More importantly, I'm not sure why you thing RMS is being used as a peak reading?
The application shows the peak, as well as the RMS and other statistics.
The RMS is not being used as a peak, it is being used as an indicator of dynamic range across the track beneath the peak

[Tone Deft]

it does not imply anything about dynamic ranges. what if the signal has a huge DC offset? there's a small dynamic range yet it can have a high RMS value.

RMS is a type of measurement, it is not a noun. yes, 'the RMS of something.' again, to ask about "the RMS" is like asking about "the blue" the blue what?

I would not look at the RMS of the output, it will not show peak levels, peak levels lead to distortion.

not picking out the OP at all... but to ask about the RMS of a track is like asking "how is the line?" how what is the line? how long is it? how wide is it? how long did it take to draw the line?

interesting though, semantics aside the idea of the original question was communicated just fine.

[Angstrom]

again, to ask about "the RMS" is like asking about "the blue" the blue what?

are you being a deliberate dick about this?
If the established subject is "the sky" then to say "the blue" is perfectly acceptable. The blue of the sky.
"Today I went out to look at the sky, the blue was incredible". Or is that also grammatically incorrect
in my post earlier I was even more specific with the subject

This will give the RMS for the whole song

that is equivalent to saying "this is the blue of the sky"
or is that not acceptable?

In my world as long as the subject of a sentence is clearly stated then we can use adjectives about it all we like

but to ask about the RMS of a track is like asking "how is the line?"

it is like asking the average height of the buildings of New York and then having someone pose objections such as

1: what if the ground is bumpy, or the building is in a hole then the average building height means nothing
2: you cannot say "the average height" .. that doesn't mean anything!
3: I would not take the average height as the peak height

all of which are willfully insistent on ignoring obvious accepted facts. Yes we check for holes in the ground when we measure the empire state building and feed it into our calculations. Yes, you can say "the average height" . Yes, we know that the average of all the buildings does not tell us about 'most tall' (that is a different question). But yes, it is meaningful to ask for the average heights of buildings in New York and compare the average to the average of London

I would not look at the RMS of the output, it will not show peak levels, peak levels lead to distortion.

lets assume that a person has rendered a file and has
1: fixed any DC issues they might have
2: resolved to export with a peak of -0.5db
3: checked the wave and it conforms to these desires.

then in this case does the RMS of the entire file contain any information about the dynamic range of the file?
or are we assuming that we fucked up the export, overloaded the output and introduced -10 db of offset and that we didn't notice it.

[Tone Deft]

wow man, not being a dick, just trying to correct some semantics.

go out into the world using this term how you want. honestly, if someone phrased it like this to me in real life I'd laugh and not take them seriously. you're making excuses for poor use of the word.

it can cause confusion if used incorrectly.

YMMV, I don't care.

edit - do you want me to correct your post? do you want to continue this?

[tw1nstates]

it does not imply anything about dynamic ranges. what if the signal has a huge DC offset? there's a small dynamic range yet it can have a high RMS value.

RMS is a type of measurement, it is not a noun. yes, 'the RMS of something.' again, to ask about "the RMS" is like asking about "the blue" the blue what?

I would not look at the RMS of the output, it will not show peak levels, peak levels lead to distortion.

not picking out the OP at all... but to ask about the RMS of a track is like asking "how is the line?" how what is the line? how long is it? how wide is it? how long did it take to draw the line?

interesting though, semantics aside the idea of the original question was communicated just fine.

Ahh you are misunderstanding what I was asking:

I was working on the basis that everyone participating in this thread would understand that a. peak level isn't going to exceed 0db, hence no distortion. And RMS was an easier way of saying your average RMS value for your track, taken either over a short period of time, say a few bars or a longer period of time, say the whole track.

Yes of course if you had a massive DC offset then you would have a larger RMS Value but you aren't gonna do that are you because one would hope we all have some sort of knowledge when it comes to producing tracks and aren't total f*cktards Well, most of the time for me anyway. . .

[tw1nstates]

Of course I didn't make it clear that I had assumed that the peak value would alway be less than 0db, I just assumed that everyone would work without clipping their final mixdown. . .

[Tone Deft]

it has nothing to do with going over a peak value.

[Tone Deft]

again, to ask about "the RMS" is like asking about "the blue" the blue what?

are you being a deliberate dick about this?
If the established subject is "the sky" then to say "the blue" is perfectly acceptable. The blue of the sky.
"Today I went out to look at the sky, the blue was incredible". Or is that also grammatically incorrect
in my post earlier I was even more specific with the subject

This will give the RMS for the whole song

that is equivalent to saying "this is the blue of the sky"
or is that not acceptable?

In my world as long as the subject of a sentence is clearly stated then we can use adjectives about it all we like

but to ask about the RMS of a track is like asking "how is the line?"

it is like asking the average height of the buildings of New York and then having someone pose objections such as

1: what if the ground is bumpy, or the building is in a hole then the average building height means nothing
2: you cannot say "the average height" .. that doesn't mean anything!
3: I would not take the average height as the peak height

all of which are willfully insistent on ignoring obvious accepted facts. Yes we check for holes in the ground when we measure the empire state building and feed it into our calculations. Yes, you can say "the average height" . Yes, we know that the average of all the buildings does not tell us about 'most tall' (that is a different question). But yes, it is meaningful to ask for the average heights of buildings in New York and compare the average to the average of London

I would not look at the RMS of the output, it will not show peak levels, peak levels lead to distortion.

lets assume that a person has rendered a file and has
1: fixed any DC issues they might have
2: resolved to export with a peak of -0.5db
3: checked the wave and it conforms to these desires.

then in this case does the RMS of the entire file contain any information about the dynamic range of the file?
or are we assuming that we fucked up the export, overloaded the output and introduced -10 db of offset and that we didn't notice it.

this is just a long list of excuses to use a term incorrectly. all these assumptions on context will screw you over sooner or later.


still, you shouldn't ask about the RMS peak value, you can still clip with an RMS peak value under 0dB. the whole point of monitoring the output is to check against clipping.

[tw1nstates]

???

I am quite confused now then,

You said that

"I wouldn't use RMS as a peak indicator, it's not the absolute peak level it's some level below that averaged over time. peaks cause clips, so you need to watch for absolute peaks on you meters."

Clipping is caused by your levels going over 0db (or whatever the absolute 0db is on your equipment as some manufacturers ill give you a little headroom on top (with gear rather than software).

Clipping is only going to be caused 9assuming that we are talking about a normal setup here, not something witha distortion unit patched into it) if your peaks exceed 0db, distorting the waveform.

Anyway, more to the point, I wasn't talking about using it as a peak indicator, rather to give an average volume level for the track, hence being able to compare it to other stuff and make sure that my mix was as hot as other stuff people were putting out.

[Tone Deft]

I get it.

[tw1nstates]

again, to ask about "the RMS" is like asking about "the blue" the blue what?

are you being a deliberate dick about this?
If the established subject is "the sky" then to say "the blue" is perfectly acceptable. The blue of the sky.
"Today I went out to look at the sky, the blue was incredible". Or is that also grammatically incorrect
in my post earlier I was even more specific with the subject

This will give the RMS for the whole song

that is equivalent to saying "this is the blue of the sky"
or is that not acceptable?

In my world as long as the subject of a sentence is clearly stated then we can use adjectives about it all we like

but to ask about the RMS of a track is like asking "how is the line?"

it is like asking the average height of the buildings of New York and then having someone pose objections such as

1: what if the ground is bumpy, or the building is in a hole then the average building height means nothing
2: you cannot say "the average height" .. that doesn't mean anything!
3: I would not take the average height as the peak height

all of which are willfully insistent on ignoring obvious accepted facts. Yes we check for holes in the ground when we measure the empire state building and feed it into our calculations. Yes, you can say "the average height" . Yes, we know that the average of all the buildings does not tell us about 'most tall' (that is a different question). But yes, it is meaningful to ask for the average heights of buildings in New York and compare the average to the average of London

I would not look at the RMS of the output, it will not show peak levels, peak levels lead to distortion.

lets assume that a person has rendered a file and has
1: fixed any DC issues they might have
2: resolved to export with a peak of -0.5db
3: checked the wave and it conforms to these desires.

then in this case does the RMS of the entire file contain any information about the dynamic range of the file?
or are we assuming that we fucked up the export, overloaded the output and introduced -10 db of offset and that we didn't notice it.

this is just a long list of excuses to use a term incorrectly. all these assumptions on context will screw you over sooner or later.


still, you shouldn't ask about the RMS peak value, you can still clip with an RMS peak value under 0dB. the whole point of monitoring the output is to check against clipping.

The peak value is the peak value, the RMS value is a *different* value. So when he is talking about a peak value of x that is the highest level for the track, the rms is the average over a period of time. 2 different things. . .

There is no RMS peak value for the context of this discussion (although if you took a bunch of RMS readings at different points in the track you could then have a RMS peak value, but that's irrelevant to what we are talking about)

[Tone Deft]

why are you trying to explain this to me? I get it. it's your wording that's all screwed up.

[Angstrom]

wow man, not being a dick, just trying to correct some semantics.
go out into the world using this term how you want. honestly, if someone phrased it like this to me in real life I'd laugh and not take them seriously. you're making excuses for poor use of the word.
it can cause confusion if used incorrectly.
YMMV, I don't care.
edit - do you want me to correct your post? do you want to continue this?

yes, I want to continue this
I am saying that: a wave which has peaks below 0db and where there is no DC offset (IE a normal wave), then in only that case I say that the RMS of the song amplitude is a good indicator of dynamic range.

So, your objection seems to be to the phrase "the RMS for the song amplitude", and you not only think this phrase has no meaning and that RMS is a 'word' which is also an adjective, but that this grammar invalidates everything else I said.


You keep repeating "the peak value at the output" stuff, yet that is not at all what is being discussed.
Also your mention of "assumptions on context" are frankly insane. We know what the context is if we perform the measurements ourselves.

The context is this, I render the wave out of ableton.
I load the wave into Soundforge and check for any flaws including clips at 0db, etc.
I select the whole wave and I collect statistics about that 4 minute long wave, which I know has no flaws
Result : I can see that my wave, with a correct peak and no DC offset, has an RMS level across those 4 minutes of -15db

explain to me why, in that specific case, the RMS level is meaningless