Should I care about RMS level?

CMolena

CMolena

Active member
Hi everybody.

I have this song that I'm mixing that sounds great, but when I watch the RMS level, in certain parts of the song, it just boosts up a lot, and I cant notice that boost in volume through my speakers. That annoys me a lot. Should I care about it, or should I just care about avoid clipping and thats it?

Thanks in advance, folks!

Sorry if I say anything stupid or obvious.
 
Last edited:
What does it sound like? It's not really RMS that causes the problem you are seeing, it's just how many meters seem to have rubbish ballistics. Back in the days of meters that moved, this was really important because you needed a peak and an average - for very different things. The BBC put's entire faith in PPMs, because transmitters didn't like peaks, and neither did recorders, so their engineers learned to read these meters. At the same time, other people were introduced to VU meters, and you often squashed the peaks, and kind of learned to work to the meter's sluggishness.

Nowadays, we have meter ballistics that seem to encourage you to avoid the top bit - your average levels are quite low on the display, but suddenly burst into the top area and scare you. You seem to have two options, work close to the top and worry about the peak light suddenly coming on, or work much lower where the levels hardly register People obsess about levels. Distortion is bad. Noise is bad. However, once out of the noise, even visually very low signals are solid and pretty well noise free, yet they seem to only just tickle above the noise floor.

I've never really seen the point in having an inch of the available meter range taken up with a raging river of the noise floor - I'd prefer this to be adjustable off the display, so anything I see is usable signal - yet everything nowadays shows you the entire range from nothing to maximum. If your monitor system can let you hear the noise at the bottom, then use your ears and just use the good bit. I get really fed up with shoving a fader and seeing a meter level go up near the top end, and not actually hear it because everything is just too high, and the range so compressed it makes no sense. What you are doing is what the meters encourage you to do - spend your time watching, not listening!
 
Yup. At this stage go with your ears and don't worry about meters and stats. As long as your meters don't show clipping you can happily ignore them (assuming some form of PPM).

Frankly, RMS measurements only become relevant when, as rob aylestone hints, you have to be concerned with what your signal is going to do to transmitters or satellite uplinks or what ever. Even then, broadcasters will probably have signal processing in their output rather than try to make every production right.

Once you'v
 
absolutely!
\

the difference between your RMS and PEAK, is your crest factor, which equals DYNAMICS
 
Thanks for the replies!

I'll try to explain the problem with more details...

The song is pretty much okay (-12db, peaking -10db) and when the guitar solo comes in...it just bursts the RMS meter (it goes up to about -7db,-6db).

When the guitar solo ends, it comes back to normal. My doubt is...should a part of a song, influence that much of the amplitude of the song?

But that being said...even with that much difference between the two parts, when I listen to it, it doesnt sound that loud, and the peaks never actually clips.

But it just bugs me..seeing the meter going crazy like that, especially because, most professional produced songs that I listen to dont do that!
 
you either have a problem with your tracks,
or your mixing.

control of level, is part of mixing.
 
A 5 or 6 dB change in level is NOT that big and certainly not an outrageous dynamic range. To the human ear, a 6dB change comes across as quite a small change.

The reason commercial recordings don't show the same range is that, after the mix is finished, the mastering engineer has compressed the hell out of it which eliminates the dynamic range.

From what you say, you're still comfortably below clipping and have room for some mastering later. Don't sweat what the meter is saying.
 
Rob, your beef about DAW meters showing the whole range?
Even way back to the (free!) Samplitude SE8 you could customize the scale to 90,60,30 even 10dB! Sam Silver ProX seems to have dropped the practice but Silver Cloud has it.

It would be unlike those clever Reaper peeps to miss this one! I bet most DAWs have the facility if you dig deep enough?

"levels" are a bit moot atmo'? Son is here and has done some USB recording using the Core 40 ID. Because he wanted to keep the acoustic levels low the resultant tracks fetched up at neg 30 or so but Normalizing boosted them cleanly to a -20dBFS average.

Dave.
 
The goofy thing about rms meters is something like a guitar solo, sine wave synth patch, or anything that is steady in the mix will run the rms level on the meter up without necessarily adding much volume to the song. Even though the rms meter is the closest meter we have to measure percieved loudness, it isn't perfect and can be tricked into reading differently than we perceive.
 
Farview said:
The goofy thing about rms meters is something like a guitar solo, sine wave synth patch, or anything that is steady in the mix will run the rms level on the meter up without necessarily adding much volume to the song. Even though the rms meter is the closest meter we have to measure percieved loudness, it isn't perfect and can be tricked into reading differently than we perceive.
Click to expand...

Another goofy thing about "rms" is that it is a bad label to put on an ever changing waveform! Since the rms value of a sine, triangle and square wave are all different how can such a value be assigned to a music waveform?

It can't of course and is no more "rms" than was the number indicated on my old Avo 8! RMS is a statistical choice. Take 3 AC meters supposedly "true" rms and feed them white or pink noise and you will likely get 3 different readings!

When talk is of keeping signals at -18dBFS "RMS" we really mean the average, i.e. a bit above, a bit below.

Dave.
 
if you watch RMS meters...

what they can inform you of, is whether or not you have really dynamic tracks.

pay attention to the difference between the RMS peak, and the PEAK...
this is your crest factor.

for CERTAIN TRACKS....
this is desirable.

for others, it is the devil.

can you imagine which tracks i'm referring to?

do you know how to get your peaks closer to your RMS, so that you have a more 'compact' and mixable track?
 
Farview said:
The goofy thing about rms meters is something like a guitar solo, sine wave synth patch, or anything that is steady in the mix will run the rms level on the meter up without necessarily adding much volume to the song. Even though the rms meter is the closest meter we have to measure percieved loudness, it isn't perfect and can be tricked into reading differently than we perceive.
Click to expand...

How do you use RMS meters?
 
dobro said:
How do you use RMS meters?
Click to expand...
In Sonar one of the choices is 'Peak + RMS'.
To me it's close enough to 'average for roughing in the body of most signals (not drums of course) in around the -20 target range, plus the eye on how the peaks are doing.
 
You must log in or register to reply here.
Back
Top