DIY preferred RMS level??

[Jouni]

Hello.


I'm just starting to doodle around about lifting RMS levels at the end of my mixes.. Doodling to get CD:s to give out to friends, bandmembers etc...

What's the RMS-level in let us say, Metal, in mastered tracks generally??


I could prolly reach some -12 to -7db with relatively listenable result, Analyzed by Wavelab... I think.

Where should I set the target, at reason??

 

[mshilarious]

-12 is probably conservative for metal. For a home job, -7 is probably pushing it. I would suspect the audio would start to break down above -9. I don't think anybody would complain -9 wasn't loud enough. Having said that, a punchy -11 could be better . . .

 

[Jouni]

Hmm.... I got -11 straight out of a mix....
With some compression an EQ...? ..You're shitting me?...

Put that side by side with an mp3 of a metalsong, and it's BURIED.

...non commercial, but my friends shouldn't have to max their stereos to hear our stuff.


I mean no offense at all!... but my -10db:s are getting buried, and that's killing it with compression from me... training, more training.... Kinda wanna know when I could proudly present the cd:s..

 

[RAMI]

Hmm.... I got -11 straight out of a mix....
With some compression an EQ...? ..You're shitting me?...

Put that side by side with an mp3 of a metalsong, and it's BURIED.

...non commercial, but my friends shouldn't have to max their stereos to hear our stuff.


I mean no offense at all!... but my -10db:s are getting buried, and that's killing it with compression from me... training, more training.... Kinda wanna know when I could proudly present the cd:s..

Are you sure you're talking about AVERAGE level for a song??? If your -10 is supposedly getting buried, there is something very wrong, either in the terminology being used, or the method.

I don't think ANYTHING goes up to -7, in any style of music.

 

[mshilarious]

Hmm.... I got -11 straight out of a mix....
With some compression an EQ...? ..You're shitting me?...

Well, one issue to be aware of is sometimes RMS measurements are different between programs. But I use WL too, and if you use the defaults, then we are talking apples to apples.

You have to realize white noise is -5dBRMS, so if you hit -7, your music starts to become barely distinguishable from white noise . . .

Which you should consider, because white noise sounds painfully bright. Probably you don't want painfully bright, you want loud. But if you increase RMS until you distort, that distortion is high-frequency energy, which gets you great RMS stats, but sounds real bad.

So if your tunes are getting lost, it could be more of an EQ balance issue than just a simple RMS measurement. I find when I start on a track, first when I fix the EQ, often the track will lose 2dBRMS. Then it's ready to turn back up to higher levels. That's what I meant when I said punchy loud is better than simply hitting a number.

But as is often said, talking about music is like dancing about architecture, so . . . post a clip!

 

[Massive Master]

WaveLab's "stock" settings read around 3dB hotter than just about any other program I've used. A -10dBRMS file (which, IMO, is ridiculously loud and bad sounding, but that's for another thread) in WL reads -7dBRMS. Maybe it's reading mono energy.

Anyway -

You aren't making your friends abuse their stereo systems - You aren't overdriving their stereo systems as bad as others are.

This question used to go around a decade ago and most people said "Aww, it'd be ridiculous to apply a "standard" to finished levels." Now, a decade later, it's gone SO out of hand, I wish the "Powers that Be" would've established some sort of "reasonable" volume -- Perhaps -15dBRMS. I've never (*NEVER*) heard a project at -12 that didn't sound considerably better at -15dBRMS.

 

[danny.guitar]

Some songs will still sound louder than other songs with a higher RMS level. So it's really hard to set a standard and I think it's kind of stupid to try and set one anyway.

If you can't hear a song with -11 on your stereo turned up then something is wrong with your stereo or your ears (possibly from listening to too many songs with -7 RMS levels ).

 

[cello_pudding]

i play folk...the loudest parts are like..-13 maybe 12.

i don't really care. if you can turn it to a considerable and listenable volume on people's stereo systems without cranking, then that's the goal. a good recording isn't about how loud or close to radio standard you can get.

and i think a good rms for getting really loud, or just listenable, can be 17...15...13..11 if its more rockish and loud.

if you want it louder, get a better stereo system.

if you want a more pure sound..stick with 13-15..or higher

 

[Jouni]

Well, here is clips... Though, this much agumenting sort of makes me think I've got it wrong somewhere. Apologies in advance.



this is mixed:

Reads some (average) left -10 and right -12db on wavelab.

This I "loudened":

reads -8.7 and -9.9db average.

OK... there's something clipping on the left.. but that , I think, is from the mix to start with...

Now, There is a difference in db, but it isn't quite audible, and I think that's lack of skill...

BUT my point is, that if played in turns(clipping or not) with a metal mp3 (Marduk-imago mortiis:L-9,9 R-9,3 average) these are redicilously quiet.

OK.. What you said, RMS isn't the whole truth. I need to dig up some info on what makes LOUD loud?....

 

[cello_pudding]

phew...i was hoping this would be instrumental rock

 

[ecktronic]

For hip hop I go for around -11 -12
Rock and pop I go for -9 -10
This all depends on how much limiting the mix can take though.
For Jazz Id go for a nice -15 or so

Eck

 

[ecktronic]

Are you sure you're talking about AVERAGE level for a song??? If your -10 is supposedly getting buried, there is something very wrong, either in the terminology being used, or the method.

I don't think ANYTHING goes up to -7, in any style of music.

Queens of the Stonage -6dB
System of a Down -6dB
Terra Diablo -8dB

Some people are just plain greedy.
Eck

 

[cello_pudding]

ok...i'm on a laptop, so consider that.

in the quieter one, i thought the crash on the 1 beat was distracting me, one the louder one, the crash was set back in the mix slightly.

what i was waiting for was that solo, because i thought the tone would be a little crappier in the 2nd. which it was...but that may have been the laptop, but...if you're going to have this on myspace or online, you'll probably want to consider your listener's speakers. especially when many people rip their songs to their computer, ipod, etc...

i would say stick with the lower one, but tame the crash just a little bit.

 

[ecktronic]

WaveLab's "stock" settings read around 3dB hotter than just about any other program I've used. A -10dBRMS file (which, IMO, is ridiculously loud and bad sounding, but that's for another thread) in WL reads -7dBRMS. Maybe it's reading mono energy.

I use Wavelab religiously for mastering and the RMS readings are the same as in Cubase.
I get a reading of -6dB for Queens of the Stonage.
It reads a seperate RMS for each channel so it is dual mono reading. I just make an average of the 2 readings.
Maybe you have it set up different in Wavelab or something.

Eck

 

[cello_pudding]

Queens of the Stonage -6dB
System of a Down -6dB
Terra Diablo -8dB

Some people are just plain greedy.
Eck

, indeed.


i know my rms meter has different ways of analyzing. what is the standard setting for it. i've got the free voxengo vst one.

 

[ecktronic]

Well, here is clips... Though, this much agumenting sort of makes me think I've got it wrong somewhere. Apologies in advance.



this is mixed:

Reads some (average) left -10 and right -12db on wavelab.

This I "loudened":

reads -8.7 and -9.9db average.

OK... there's something clipping on the left.. but that , I think, is from the mix to start with...

Now, There is a difference in db, but it isn't quite audible, and I think that's lack of skill...

BUT my point is, that if played in turns(clipping or not) with a metal mp3 (Marduk-imago mortiis:L-9,9 R-9,3 average) these are redicilously quiet.

OK.. What you said, RMS isn't the whole truth. I need to dig up some info on what makes LOUD loud?....

What makes a mix sound loud is lower RMS value and more mid range frequencies in your mix. The ear is most sensitive to the high mids.
PLay about with frequencies in the range of around 1k to 6k
1-2k 3k and 5k are good staring points for boosting. Watch for harshness though, This is where the quality of your recordings comes into play. Bad quality recordings could sound very harsh in these frequencies.

Eck

 

[Jouni]

I just got wavelab, it's in it's defaults.
..and I know shit about it or anything else.

Well I think, the low end is eating up alot of energy in my stuff. (bassist )
(might be missed on a laptop?)

My guitars, I like middly, during mix I usually end up cutting heavily on 400hz and adding stuff in 1k-3k. Most on leads.

1k-6k huh?... I'll keep that in mind, also remembering that RMS around -9 is adequate.

The differences in style/rms would be explained by number of hits/time??
In slower tempo/intensity music, the average RMS is reduced by more silence and less hits per song??...

 

[mshilarious]

WaveLab's "stock" settings read around 3dB hotter than just about any other program I've used. A -10dBRMS file (which, IMO, is ridiculously loud and bad sounding, but that's for another thread) in WL reads -7dBRMS. Maybe it's reading mono energy.

No, because it gives separate left and right channel reading. I think it is calibrated accurately too, because a full scale sine wave should be 20*log(.707), or -3dBRMS, and it is.

Also, the K-meter in the UAD Precision Limiter matches the WL RMS, at least for test waves. I don't have WL6, which now has internal K-metering.

Maybe there is some difference in program material


The loudest thing I ever heard that I thought sounded good was a clip Mixerman posted a while back here. It was metal, loud as heck and so punchy I think I peed myself. That was a rough mix as he described it, better than anything I've ever thought about finishing; I think it was -8 or -9.

I bet a lot of major label stuff sound that good at mixdown, until it is killed in mastering

Anyway, asking why you can't hit -7 and sound good is like me asking why I can't drive 300 yards if Tiger Woods can. Well, truth be told, I can hit 260 on a good day with a tailwind, but when I try to hit it that hard, half the time I end up in the fairway . . . of a different hole

 

[mshilarious]

OK I am listening to your tracks. My first impression is that the style isn't really like a loud metal track. The guitars have a more bluesy tone, not that in-your-face distortion. It's a more classic style of metal than the new superloud stuff.

As for balance, it seems like the kick needs to hit a lot harder, get that fundamental at 60 Hz up. The guitar tone is tough, it's a little bright, a little boxy, so try to work with that to get a more open sound.

I took a quick shot, I think this can work in the -9 range, but it needs lots more balls to sound like it is loud. It's not as easy cranking the limiter. At -8, it starts to fall apart, at least with me at the driver's wheel.

 

[RAMI]

Queens of the Stonage -6dB
System of a Down -6dB

I'm not saying I don't believe you. But I find it so hard to believe that I would have to see it for myself. I'm not so sure I'd want to hear it, though.