getting volume to commercial cd levels?

[SouthSIDE Glen]

K-14 to K-12 FS. Mixes have to end up somewhere and that's where I work to.

Regardless of content and regardless of mix quality?

Just playing devil's advocate here, but for example; if you have wall of sound metal mix that is coming out of the two mix at -14, does that mean you leave it alone or maybe squeeze another 2dB out of it? Or if you get one that cones out of the mix at -11 that you strap an expander to it?

Or if you have an acoustic mix that comes in at -19, do you automatically squeeze 5 to 7 dB out of it?

And finally, do you target those levels pre or post spectral sweetening?

Again, this is not meant as an attack if even the questions seem somewhat pointed. I'm just trying to understand the technique is all.

Because frankly, the way I'm used to doing it, setting target numbers first is kind of switching around cause and effect. I'm used to letting the content dictate the process, the numbers will fall where they may. I can easily create - and often do - a self-mastered CD where the individual RMS levels for each track can vary anywhere between -17dBRMS and -12dBRMS - with those numbers simply being the result of responding to the content and not predetermined values - and still provide a consistant listening experience for the listener, exactly because of the differences in content. And I just don't understand how to acheive that same consistant quality product by targeting a more-or-less arbitrary mastering level.

Perhaps we deal with different styles of artist/client. Maybe my regulars produce more variation in content style and arrangement with a wider mix between sparse ballads and dense anthems. Or maybe you just have a trick up your sleeve that I have not got a handle on yet. That's what I'm trying to figure out.

EDIT: What I suspect, but am honestly not going to say for sure just yet, is that a big part of the answer may be because there is so much compression already applied to the mix. Which would indeed take us back to the whole question of mix quality and it's effect on mastering.

G.

 

[Keiffer]

Regardless of content and regardless of mix quality?

Just playing devil's advicate here, but for example; if you have wall of sound metal mix that is coming out of the two mix at -14, does that mean you leave it alone or maybe squeeze another 2dB out of it? Or if you get one that cones out of the mix at -11 that you strap an expander to it?

Or if you have an acoustic mix that comes in at -19, do you automatically squeeze 5 to 7 dB out of it?

it depends...

playing devil's advocate... do you arbitrarily set levels with no game plan? do you not have a preferred set of tools that you generally consider your starting point?

And finally, do you target those levels pre or post spectral sweetening?

I've done both and neither. but never after the final limiter.

Again, this is not meant as an attack if even the questions seem somewhat pointed. I'm just trying to understand the technique is all.

Because frankly, the way I'm used to doing it, setting target numbers first is kind of switching around cause and effect. I'm used to letting the content dictate the process, the numbers will fall where they may. I can easily create - and often do - a self-mastered CD where the individual RMS levels for each track can vary anywhere between -17dBRMS and -12dBRMS - with those numbers simply being the result of responding to the content and not predetermined values - and still provide a consistant listening experience for the listener, exactly because of the differences in content. And I just don't understand how to acheive that same consistant quality product by targeting a more-or-less arbitrary mastering level.

Perhaps we deal with different styles of artist/client. Maybe my regulars produce more variation in content style and arrangement with a wider mix between sparse ballads and dense anthems. or maybe you just have a trick up your sleeve that I have not got a handle on yet. That's what I'm trying to figure out.

G.

this can be debated ad nauseam. that's good info though. that was my goal... to provide info and insight. thanks.

rockabilly1955, I hope my initial post helped.

kp-

 

[kylen]

... I can easily create - and often do - a self-mastered CD where the individual RMS levels for each track can vary anywhere between -17dBRMS and -12dBRMS - with those numbers simply being the result of responding to the content and not predetermined values - and still provide a consistant listening experience for the listener, exactly because of the differences in content.
G.

Yes - that's kinda what I have in mind when I might mention K-12 somewhere...meaning K-12ish at the loudest moderately dense rock track. I don't really like the sound when it gets louder in general but the hottest track might go into the 11's - I think I can tell when it goes into the 10's - on my equipment anyway (local remastering guy here!). On Bob Ludwigs equipment it can go down to K-6 and sound acceptable for a while - don't know how he does that one! I think equipment plays a fair percentage in the equation in this case.

 

[SouthSIDE Glen]

playing devil's advocate... do you arbitrarily set levels with no game plan? do you not have a preferred set of tools that you generally consider your starting point?

Keep in mind that I am not an ME, nor do I play one on the Net. I'm just a mixing engineer who often is called upon to self-master the mixes he does. But here's how I do it, more or less:

If I'm doing a single song, I do not "set" my levels anywhere. I edit the dynamics and dial the knobs by ear, taking the song up until the mix begins to fall apart, and then back off of that just a taste, normalling my peaks peaking no higher than -0.1dBFS. Limiting reduction on the mix is typically 2dB or less, not by design but by the way it usually just works out.

If I am doing a compliation of songs for EP or CD destination, then I'll sample all the songs, and perform the above procedure on the quietest (RMS), least dense of the songs first. I then use that as an aural baseline for the rest of the disc, mastering the other cuts to aurally fit with the quiet one. Some of the stronger anthems might get a little extra juice anyway, but it depends upon the layout of the songs and how I set up the rhythm of the CD as a whole.

I usually watch or check the numbers out of informational curiosity and as a litmus test to make sure I'm not making any major errors in judgement - if a song winds up 3dBRMS different than it's predecessor but they are of similar content and arrangement, I might double-check a few things just to make sure I'm not fatiguing or forgot to push a button somehwere. But I never "master by numbers", in that I never decide beforehand, "OK, this one needs to be at -14", and I accept that 3dB difference if I confirm that I have not made a mistake, it just winds up sounding best that way, and certianly don't expect or want them to all come out the same.

Different songs with different natural sonic densities need to have different final RMS levels if you want them to sound like they are meant to go together on a playlist.

It's all up to the content and the mix.

G.

 

[Massive Master]

oh please... it is not... how's raising the level in the final product as you say "making up" for mixing. if that's the case then a lot of MEs produce deficient mixes.

in this case... if half the time was devoted to talking specific technique. that was the question. this is home recording. technique specifics are important.

kp-

I'm saying that if you need to chain four dynamics controllers in a chain to get to "commercial levels" than the mix didn't want to be there in the first place.

It's a piece o' cake to get a good mix to "commercial levels" simply by throwing a ("a'" as in "one") decent limiter across the bow and tweaking the A&R.

Okay, that's somewhat of a generalization, but it rarely takes much more than that. A mix that has proper dynamics control *by the talent* then by the mixing engineer, usually doesn't need extraordinary measures taken to get to a reasonably "commercial" level at the mastering stage.

And believe me - I understand this is home recording. That's why I try to get to techniques that produce good recordings. Not loud recordings.

As much as I hate the volume trend, most good recordings have no problem being loud recordings. Unfortunately, half the time (or more) the people who are trying to "make their recordings loud" don't even understand how to *track* their recordings properly. I'd rather get them to concentrate on that. Without the basics, everything else is putting the cart before the horse.

 

[masteringhouse]

FWIW I use multiple dynamic processors quite frequently to control different aspects of the dynamics or to add a particlular color.

One of the nice things about the Weiss DS1 is that it has several release times built into one box. Without a unit like that multiple processors make sense if the material needs it. I'll often use my first compressor to control some of the transients of the drums and to make them a bit "meatier" (faster attack/release with threshold set high). Next in the chain might be one for helping to obtain a better average level and to add a bit of "glue" to the mix (slower attack/release, threshold deeper in the mix). A third (often the Weiss) starts to get into loudness territory where it may be used as a front end to the L2 so that the L2 isn't working quite as hard, and lastly an L2 for volume, maybe a 1-2 db increase if the client doesn't twist my arm into hypercompression

Spreading out the workload like this helps to create volume without as many artifacts. Note that each comp in the chain is only a mild compression though. For example the Weiss may only be at a 2:1 ratio or lower with only a few db of reduction on each comp. It's when you strap on a single limiter and expect it to do all of the work that I think many fall into trouble.

 

[CIRO]

I'm saying that if you need to chain four dynamics controllers in a chain to get to "commercial levels" than the mix didn't want to be there in the first place.

It's a piece o' cake to get a good mix to "commercial levels" simply by throwing a ("a'" as in "one") decent limiter across the bow and tweaking the A&R.

Okay, that's somewhat of a generalization, but it rarely takes much more than that. A mix that has proper dynamics control *by the talent* then by the mixing engineer, usually doesn't need extraordinary measures taken to get to a reasonably "commercial" level at the mastering stage.

And believe me - I understand this is home recording. That's why I try to get to techniques that produce good recordings. Not loud recordings.

As much as I hate the volume trend, most good recordings have no problem being loud recordings. Unfortunately, half the time (or more) the people who are trying to "make their recordings loud" don't even understand how to *track* their recordings properly. I'd rather get them to concentrate on that. Without the basics, everything else is putting the cart before the horse.

MY SONGS
http://www.soundclick.com/openstation

 

[kylen]

...that was my goal... to provide info and insight. thanks.

Yeah thanks ! I've really thought a lot about what everyone had to say and went back and attacked a problem remaster that wasn't working out earlier - and have impressed myself (at least!) using different approaches and techniques mentioned here!

...except the one using the Weiss

 

[danny.guitar]

I'm no engineer, or professional, but I have to agree that the track needs to be good to start, this is just from experience. Use a compressor if you need to, to control the peaks, however, since I usually record solo acoustic guitar, this isn't much of an option for me, so recording quality tracks is the only 'fix' in my situation.

After that, as far as volume goes, I usually just normalize to about -2db. This is just what sounds good to me. If you have lots of peaks, normalizing wont do much good, since only the peaks will be at -2db, the rest will be relative, and still much lower in volume.

Either way, I dont usually worry about volume levels until I've finished a number of different songs, so I can make them at about the same volume as other songs I record, so they sound right on CD...that way one song isn't louder than the other.

Maybe my method seems over-simplified, but that's just what I do.

 

[Keiffer]

Let me clarify... the reason for using up to four compressors is to spread the compressing load. several doing mild and moderate compression in series can be much more effective than one carrying the full load.

 

[kylen]

Let me clarify... the reason for using up to four compressors is to spread the compressing load. several doing mild and moderate compression in series can be much more effective than one carrying the full load.

OK - I got it, good approach. Precedents in RNC compressor and mentioned in "Mastering Engineers Handbook" [Osinski], etc.You push one peice till it sounds it's best, then instead of pushing it higher if you need more throttle you add a second piece to push farther and so on maintaining the best and most appropriate sound throughout.

[edit: Oops forgot to mention masteringhouse above, obiously the technique works for him listening to various samples of his work!]

 

[masteringhouse]

Let me clarify... the reason for using up to four compressors is to spread the compressing load. several doing mild and moderate compression in series can be much more effective than one carrying the full load.

Exactly! As I mentioned above.