Acoustic RMS Level

Just finished mastering four acoustic CDs (vocal solo and acoustic guitar only) and I'm not exactly sure where to set my final levels.  Obviously, I want to preserve the dynamics, but they need a little more "push" than they currently have (current rms levels range between -20 to -22).  I'm thinking maybe they should be somewhere around -15, but I'm not exactly sure what commercial industry standards are (if any??) for this type of project.  Any suggestions? 

I could be confused but if the mastering is finished, doesn't that says you are not considering changes to the mix such as compression or limiting, only doing simple amplification? Is there any downside to bringing the peaks up somewhere near maximum?

I can’t say about industry standards but a few acoustic music CDs I’ve looked at out of curiosity have had their highest peaks near 0dB, often making for a 15dB or more difference with what came immediately before and after. I never made RMS measures on them as far as I can remember, but it was visually apparent that it had to be well below those peaks.

Those CD were undoubtedly more than 10 years old. The general music industry standard these days seems to be something like: push the peaks to 0 and beyond, and make sure the average isn’t more than a few dB lower.

I'd try to find a reference track and measure the RMS on that.

James.

I could be confused but if the mastering is finished, doesn't that says you are not considering changes to the mix such as compression or limiting.

Basically, the mastering is done -- but still considering a little amplification/limiting.

If you want a natural, acoustic sound, you may possibly want to leave it alone.  As a classical guy, I could care less what the RMS level of a CD is...  if it's not loud enough, I turn up the volume.  But in the context of a pop world, I understand that shuffling tracks can be quite troublesome with a recording such as this. 

Since the CD is already mastered, I'm presuming that it is normalized.  So how much time will be affected by raising the volume and adding peak limiting?  If it's one or two brief moments out of a whole CD, that added compression might be a reasonable tradeoff.  If many peaks would have to be reduced, my bias is always to leave it alone.

In looking at a waveform, quite often I've found that sharp peaks can be quite numerous, but maybe not intended. All it takes is for a guitar string to be plucked a bit too hard, and certain vowels in words to mess with the mic diaphragm, or simply a vocalists poor performance, and that can happen enough to keep the bulk of the performance down to a volume that is not ideal.

I control all that so the ear isn't always being tampered with during listening. And if vocals are involved, just perhaps you may lose some intelligibility because the peaks keep the softer vocals too low. Some judicial (artistic) compression up at the top followed by extremely light limiting, at the top, can smooth a final product to perfection.

I wouldn't shoot for a specific rms level, but at least control aberrant peaks. Let your eye tell you what needs to be done. Then check it with the ear. Once that's done, the rms level will find it's own perfect place for the recordings involved.

I control all that so the ear isn't always being tampered with during listening.

Hmm... and there's me thinking that by definition, listening to sounds involves tampering with the ears!

Thanks to all for the input.  I wasn't necessarily looking for a specific number -- just a general range.  Obviously, this type of material would be very different from a hip-hop, rap or heavy metal project.  Thanks again.

I control all that so the ear isn't always being tampered with during listening.

Hmm... and there's me thinking that by definition, listening to sounds involves tampering with the ears!

Well -- yes, by cracky, you just may be correct, but fortunately I feel secure in knowing you know what I mean, though semantics can be a fun carousel. I'll concede I left out the word 'unpleasantly', as in 'tampered with unpleasantly'. Or, maybe more appropriately since I'm from the USA - "so in listening the listener doesn't get yanked around". There, that's definitely more regional.

Gotta love it!

Another option for raising overall level, often with minimal aural impact, albeit time consuming, is to look at the audio wave forms in your editing program and lower in volume just the loudest peak(s); if one stands out 3db above the rest, lower it and bingo, re-normalize your project 3 db louder.

Very true -- and with the mastering limiter in iZotope's Ozone 3 or 4, you can do that automatically without having to find all those peaks yourself.  But before I had Ozone, that's exactly what I used to do, Dave.

Very true -- and with the mastering limiter in iZotope's Ozone 3 or 4, you can do that automatically without having to find all those peaks yourself.  But before I had Ozone, that's exactly what I used to do, Dave.

Well, I guess I'll keep doing it the old fashioned way... not being an iZotope customer I'm looking at $250 USD to buy in for that pleasure.     Looks good though.

I thought Audition would have included a decent look-ahead limiter by now. If it doesn't then try looking for Kjaerhus Classic Master Limiter which seems to work well and is free.

James.

Another option for raising overall level, often with minimal aural impact, albeit time consuming, is to look at the audio wave forms in your editing program and lower in volume just the loudest peak(s); if one stands out 3db above the rest, lower it and bingo, re-normalize your project 3 db louder.

Maybe I'm missing something, but isn't this the same as gently hard limiting, or using compression above the selected threshold? If so, then why look for peaks when the software will do that for you?

Another option for raising overall level, often with minimal aural impact, albeit time consuming, is to look at the audio wave forms in your editing program and lower in volume just the loudest peak(s); if one stands out 3db above the rest, lower it and bingo, re-normalize your project 3 db louder.

Maybe I'm missing something, but isn't this the same as gently hard limiting, or using compression above the selected threshold? If so, then why look for peaks when the software will do that for you?

Its just a way to control the audio with minimal aural impact.  OP wants to preserve dynamics; gently hard limiting or using compression can change the sound entirely.