Default RMS level for songs

godchuanz

New member
Hi,

I'd just like to ask all the MEs and ME-to-be's here... if a client sent you a mix with sufficient headroom, what is the default RMS level you bring the mix up to? I understand it may depend on the genre, but what's the RMS level you aim for in a Hard Rock song? What level do you aim for in a Country song? Maybe a Hip-Hop record?

On a related note, if everyone in the world had the sense to use the volume knob on their playback devices (i.e. without the loudness war), what levels do you aim for?

Cheers!
 
Hi,

I'd just like to ask all the MEs and ME-to-be's here... if a client sent you a mix with sufficient headroom, what is the default RMS level you bring the mix up to? I understand it may depend on the genre, but what's the RMS level you aim for in a Hard Rock song? What level do you aim for in a Country song? Maybe a Hip-Hop record?

On a related note, if everyone in the world had the sense to use the volume knob on their playback devices (i.e. without the loudness war), what levels do you aim for?

Cheers!

I've never looked at or went by a digital read out to shoot for or know what the highest peaking or average rms number is when judging loudness.

People will work differently, but any kind of number in the regard to loudness will be misleading and I feel it's much better just to use your ears in comparison to other tracks when you are interested in getting your tracks in the neighborhood. When going for higher levels, what can be important is not losing transient detail and sense of balance.
 
As there really isn't an answer to the first part --- I know a lot of people throw around a bunch of numbers but none of it is really going to make an awful lot of sense, I'm going to jump to the more important part:
On a related note, if everyone in the world had the sense to use the volume knob on their playback devices (i.e. without the loudness war), what levels do you aim for?
I'll step in this one because it was an early thing when I was first dragged kicking and screaming out of my comfort zone and into the mastering chair (long story that I won't get in to at the moment).

When that happened, the intense study began. And all of it was done at fixed output levels (I still feel that people have no idea how important it is to have a properly calibrated chain from start to finish, but that's for another thread). Low volume, mid volume, high volume, low end, midrange, top end, apparent compression at what point, apparent limiting at what point, how are the effects and space 'moving' (was there a compressor on the reverb buss or is it 'breathing' from the mastering stage), etc., etc., yada, yada.

One thing stood out almost immediately and hasn't changed since (and this was 1993-1994-ish). A ~15dB crest factor (lets broaden that to "14-to-16dB") tends to be where pop/rock/metal/rap (non-acoustic, non-classical, modern and contemporary recordings for lack of a better term) is happy. And to that end, I have yet to hear *any* recording with less than a 15dB crest that didn't sound better *with* a 15dB crest.

Not that I'm trying to say "universal" -- but it was close. If recordings came in with a 18-24dB crest, you could almost bet that while decreasing it to around ~15dB would add perceived "power" and "energy" and "groove" and all those descriptors that people love attached to their music. Open, dynamic, powerful, unconstrained, fast, etc.

Past that point -- I mean RIGHT past in most cases -- Start pushing the crest down to 13 or 12dB and you've killed it. Squashed, constrained, slow, unexciting, irritating, etc. You've not only removed the listener's reason to crank it up, you've removed the reward for it.

Is it a number to shoot for? No. I don't go by numbers and meters (except during the all-important calibration mentioned earlier). But it was a number that came up so frequently, nearly universally, that it was worth spending the last half-hour trying to type a reply to a forum post.


* And note that although it's easy enough to move the "crest factor" numbers into "dB(FS)RMS" numbers, I'm sticking with the former because that's what really applies. If a project calls for -24dB(FS)RMS levels, such as high-res or motion picture projects, that same crest is where the music is probably going to sit.
 
Massive Mastering,

Thanks for sharing. Great insight!

I started this post because I was quite confused by the advice of many mastering gurus - mostly discouraging the loudness war. Advice/Guidelines like don't push peaks to absolute 0dB... the limiter shouldn't have gain reduction of more than 1-2dB, or... apply gentle compression with ratio no higher than 1.5:1... Always check for inter-sample clipping... All sound like really good advice to retain dynamics, but in practice are these really being done?

A crest factor of 15dB would probably mean an RMS lower than -15dB most parts of the song. Really honestly, I struggle to even find a song in my library that has RMS levels below -12dB. Even albums widely regarded as well-mastered, when I analysed the levels, I see peaks at close to 0dB, with RMS at around -10dB throughout most of the song. Some pop recordings even push above -6dB. And I have not seen a single modern pop song that passes the inter-sample clipping test. The inter-sample clipping red button on the level meter lights up almost all the time when I play any modern song through it. Some of these songs even sound good. (Ok... maybe they will sound better with 15dB crest factor :D)

Based on my observations, most MEs push the levels to between -10dB and -7dB for Rock songs. And this include those really renowned MEs - the same few names you always see in the credits section of award-winning albums. On lighter songs, sometimes I may see around -14dB to -12dB. Frankly, if the MEs do apply the supposed best practices, it's near impossible to end up with songs this loud. It's almost like all the mastering gurus are saying one thing, but all the elite MEs are doing another. So are the top engineers all part of the loudness wars, or am I missing something?

Really hope more MEs can share their thoughts... I need to master my album and I don't wanna be part of the loudness wars, but I also don't want to end up regretting not getting help from the real experts (who seem to be all part of the wars).
 
Hi,

I'd just like to ask all the MEs and ME-to-be's here... if a client sent you a mix with sufficient headroom, what is the default RMS level you bring the mix up to? I understand it may depend on the genre, but what's the RMS level you aim for in a Hard Rock song? What level do you aim for in a Country song? Maybe a Hip-Hop record?

On a related note, if everyone in the world had the sense to use the volume knob on their playback devices (i.e. without the loudness war), what levels do you aim for?

Cheers!

Waltz and Massive gave more technical responses than I could ever do (I'm not a mastering engineer by any means), but if I was to make it pretty vague, I'd tell you that by genre, most relaxed, indie/acoustic kinda music I've seen is usually around -14 or -12dBFS RMS, most rock I've seen is closer to -10 or -8, and I've seen metal get as loud as -6 RMS.
Of course, a lot of metal masters sound god awful and full of distortion (take a listen to the latest Punk Goes Pop album... The ME did awful on that), but the numbers I'm giving you are just what I've experienced seeing if you're looking for hard and fast numbers.

That being sad, listen to the pros here. Use your ears for this, not your eyes for this kind of stuff. I REALLY like how Massive was talking about the ~15 dB crest factor sounds good, but past that it's squashed. If you take a listen to modern pop/rock/metal music, almost all of it is squashed lifeless. It WILL sound better with more crest factor, but the only issue is consumers nowadays don't know how to use those volume knobs in their car. If they're used to turning their car stereo to 16 and they have to turn yours up to 20, a lot of people will think it doesn't sound as good (which is just silly, but it's the world we live in).

Anyhoo, hope I helped a bit. Enjoy your audioz. =]
 
Based on my observations, most MEs push the levels to between -10dB and -7dB for Rock songs. And this include those really renowned MEs - the same few names you always see in the credits section of award-winning albums. On lighter songs, sometimes I may see around -14dB to -12dB. Frankly, if the MEs do apply the supposed best practices, it's near impossible to end up with songs this loud. It's almost like all the mastering gurus are saying one thing, but all the elite MEs are doing another. So are the top engineers all part of the loudness wars, or am I missing something?

Beat me by two minutes... xD

I only know of one number that has any real value in rock music..... 11

Which, if you think about it, is exactly what the binary code for a REALLY squashed track is. 11's over and over and over. Maybe that's why Rock masters are getting so loud nowadays...
 
I hardly ever look at the DAW meters for the sake of choosing a "target level" for my songs. When I edit tracks and do comps, I use the meters to make sure my edits are level-balanced, etc....but when I mix, I never bother.

I calibrated my monitors and my console a long time ago. I have 3 marks that I made on my TC Electronics Level Pilot monitor level knob --- 75-80-85 dB SPL --- these were all done using an SPL level meter. I know if I go a notch above the top mark, I'm at 90 and a notch below I'm at 70.
So I always set my overall level there, and then the individual track levels before the monitors are all done "by ear" to those reference marks.
When it's sounding good-loud against my marks...I know the metering and all other levels upstream are where they should be.
 
Based on my observations, most MEs push the levels to between -10dB and -7dB for Rock songs. And this include those really renowned MEs - the same few names you always see in the credits section of award-winning albums. On lighter songs, sometimes I may see around -14dB to -12dB. Frankly, if the MEs do apply the supposed best practices, it's near impossible to end up with songs this loud. It's almost like all the mastering gurus are saying one thing, but all the elite MEs are doing another. So are the top engineers all part of the loudness wars, or am I missing something?
I'm as guilty as the next guy of making recordings far louder than they should be. Most of us are. We don't make recordings ridiculously loud because we want to or we think they actually sound good that way... We do it "under protest" to The Boss (which is the client).

The "loudness wars" aren't the blame of mastering engineers run amok... It's a pissing contest between bands (and labels, etc.). The listener never asked for this -- Polls show the listener doesn't care about it either. And lord knows if they knew what we know (about what songs at the level they "want to be" sound like vs. what they sound like when the end listener hears them) there would be an uprising that would make the Death Magnetic protests look like a tea party.

I can't speak for them all of course -- But I can guarantee you the vast majority of us (and by "us" I mean the people who do this as a speciality - Not the guys who "Master" stuff with a capital "M" and talk about nothing but "Industry Standard Loudness" and "Industry Standard blah, blah, blah) would give (something - not sure what - perhaps a testicle or endure certain physical tortures) to go back to "normal" levels. I was asked in an interview in '94 or 95' whether there should be a "standard" level like the MPA. I (stupidly, naively, pick an adjective) said something along the lines of "Of course not. It's not like artists would wreck their recordings just to make them too loud."

I would kill (something - not sure what) to go back to the levels that "loud" was back then.

We didn't start it, we didn't want any part of it, we don't like spending $5k for a compressor just because it has 2dB more usable headroom than another. We don't like calibrating line level to -10dBFS=0dBVU or whatever some guys are using these days (my first 24-bit converters were at -20dBFS, which IMO, is a pretty nice place to be. Lower would be even better for tracking and on the rare occasion that I am, those converters are usually calibrated to -20dBFS if I have any say in it).

But we do it. Because we won't be able to eat otherwise. If it makes a difference, at least many of us do get the gear with obscene amounts of usable headroom -- We do recalibrate converters to handle levels that would make tracking or mixing impossible (or at least impractical). We do all sorts of things that "limit the damage" as much as possible because we know many clients are more than willing to do as much damage as possible just so their tune is as loud as the previous tune.

[/rant]
 
Massive Master - that is the best rant i've read on the internet in a very, very long time. And i don't like to use the word "rant" for it because it's not really; it's honest, it's passionate, and it shows you genuinely care, heart and soul, about what you do! I've always had a lot of respect for you for your knowledge and your way with words, and this just reinforces my reasons for that respect.
 
Massive Master - that is the best rant i've read on the internet in a very, very long time. And i don't like to use the word "rant" for it because it's not really; it's honest, it's passionate, and it shows you genuinely care, heart and soul, about what you do! I've always had a lot of respect for you for your knowledge and your way with words, and this just reinforces my reasons for that respect.

I agree with this gentleman whole-heartedly. What you just wrote should be in textbooks if it isn't already.
 
Thanks for some very interesting posts, Massive!

May I ask a question? When you say that most music sounds better with a 15ish dB Crest Factor, what listening conditions are you in? FYI, sitting in my studio or even my living room, I totally agree with your take on dynamic range, but....

I have a theory that one of the biggest culprits in the loudness wars was the invention of the dreaded Walkman then Discman then MP3 player. These portable devices mean that, instead of listening to music in the controlled environment of a living room or bedroom, much listening (perhaps most) is done in noisy environments likes streets, buses, trains, cars, etc.

These noisy environments have two effects. First, people crank up the volume to drown out the background noise (which, with earbuds, is very likely to cause permanent hearing damage). Second (and relevant to this topic) it means that if a recording has a normal (and, I agree, nice sounding) dynamic range, any quiet bits tend to get lost.

Taking this one step further, I think the resulting loudness wars have had another undesirable effect: since commercial recordings now have such a limited dynamic range, people have forgotten how to listen. Unless everything is at a monotonous (and relatively loud) level, a large percentage of an audience in a theatre or cinema complain they couldn't hear the quiet bits. I've seen actual studies done on this part--50 years ago people would go to an unamplified play and hear every line and nuance. Now, even with radio mics, people complain that they can't hear the quiet bits.

(My personal rant is that, with air conditioner noise, moving lights and noisy audiences, auditoriums have also become a lot louder but that's a different topic for a different forum.)

Anyway, I'm curious to hear your thoughts on changes to listening locations and conditions and their effect on acceptable dynamic range.
 
Back
Top