mastering song by song, or the album as a whole?

bkkornaker

bkkornaker

www.bryankmusic.com
heres my dilema for all your mastering gurus out there......

Lets say i have a 10 song cd (songs have been mixed), and all the songs are ready to be mastered. Would it be wiser to master each song individually? or should one master them all together?

for example....im useing Wavelabs 5, and i can either master each song entirely on its own and export the mastered song, and do this one by one for each song......

....or....should i open all the songs up in a cd montage, and put my mastering effects at the master buss, and have a general "one" mastering preset that will effect all the songs on the whole album?

i know it sounds dumb, but which would be better, and the most common approach?
 
My answer is: Neither and/or both.

You offered the choice between one at a time, or all at once with the same setup. My approach is to split the difference and combine your ideas.

1) I do them all at once. I look at mastering as a "state of mind" kind of practice. I want to be all done tracking and mixing. I want fresh ears and a fresh mind, and I want to focus on that one task; thus I master all tracks in a project at once.

2) I don't apply the same effects (limiting, compression, eq or whatever) to them all. Besides some of the standard reasons for mastering (more punch, more pop, more volume--yeah, all the same thing) for me, the real value in mastering is pulling the individual tracks in the same direction--so they sound like they belong the same CD.

I don't know about you, but the way I track sometimes, by the time I'm done with a longer project, the approach is likely to change. So to give the tracks some consistency--some glue to hold them all to the same sound--they might each need something different. One might need some high end to breathe a bit, one might be too hot, etc.

So it seems to me that only if each song were tracked in the same environment with the same results would the mastering formula be the same (and maybe that's you--but like I said, it's not me...)

For me the goal is pulling them all together, even if that means (and it often does) pulling them from different directions.

And on a completely different topic--for quite a while now, as I've been lurking and learning here, I've completely enjoyed your music. I think I can appreciate craft and art in any genre, but your style happens to be my favorite--your playing and your recording are both a blast for me to hear!
 
I am not a pro mastering engineer, but when I do pretend to be, I combine both ideas. Here's how:

Each song IMHO must be mastered seperately, however they must be mastered with each other in mind. Each song will need it's own EQ, it's own fades, it's own peak management, etc., however they will have to fit cohesively into the album, blending into each other properly and hitting relatively even perceived volume levels.

Unless one has created a very boring album, there will be variation in the density and dynamics of songs throughout the album, meaning getting balanced perceived levels will have to be done by ear, and be set differently for each song rather than picking a numeric volume level that will be applied the same for every song.

The way I handle that is to master the apparently quietest song on the album. That means the one with the least sonic density, the sparsest arrangement or mix, etc. I'll master that to sound as good as possible and as loud as it will go before it starts to get shrill or start to break up. Then I use that as the reference listening level to which I try to match the rest of the songs.

From there I word from that song outward in the album. For example, if the quietest song - and therefore my reference song - is going to be track 4 on the CD, I'll do track 3 and track 5 next, getting the timbres the volume and the fades/transitions correct. Then track 2 and track 6, etc.

G.
 
I agree with Glen except that I usually start with the loudest song and hopefully the one that best represents the "average" as far as balance. My reason for picking the loudest first is to not "paint myself in the corner" so to speak. If you take the most quiet, and bring that up to a level that you think is appropriate, then go for the loudest, you may not be able to make it loud enough in comparison without distortion.

I also automate settings for each song where possible, and jump from song to song on the entire album to hear how it holds together before creating the final versions.

Just my thing though, everyone is different.
 
masteringhouse said:
I agree with Glen except that I usually start with the loudest song and hopefully the one that best represents the "average" as far as balance. My reason for picking the loudest first is to not "paint myself in the corner" so to speak. If you take the most quiet, and bring that up to a level that you think is appropriate, then go for the loudest, you may not be able to make it loud enough in comparison without distortion.
...
Just my thing though, everyone is different.
Click to expand...
That's interesting, Tom. Could you maybe explain that a little more?

I'm NOT disagreeing, or saying that one way is right or wrong, I just don't understand, and I may be missing something important. From my viewpoint, I do the quiet one first because it seems to me that it would be just the opposite; i.e. that if one started with the loud one and brought that up to it's maximum ideal level, that it might be pushing the overall perceived volume beyond where one could comfortably push the quiet one; i.e. it might be setting the bar too high.

OTOH, by starting with the quieter one, it's kinda like picking the lowest common denominator. By getting the maximum ideal level on the quiet one and using that as the standard, it would be easy to get the louder ones sounding good and sounding compatable without having to push them beyond their limits.

Then again I'm just a part time Sears ME. I don't have the experience or the gear of a real ME such as yourself, so maybe it's my limitations. Or, maybe like usual, I'm just thinking about it the wrong way :rolleyes:.

G.
 
Well let's start with a worse case (and often a little too familiar). The loudest song is crushed within an inch of it's life. There is maybe 2 db that you can work with. At that point the overall RMS level for the CD is already pretty much set and you can't make much if any adjustment. On the other hand, the softer songs are still something that have room to work with and can be raised to the relative level of what has already been etched in stone.

If one had started with the softer song and brought that up to the level that you feel it should actually be at, then tried to bring up the brick so that it's relative level was appropriate in comparison to say a softer ballad, you're screwed (or have "painted yourself into a corner"). You will need to go back to the softer song and lower it based on what you can get out of the louder track.

I guess what I'm saying in a nutshell is first set your parameters on the things that you can't change.
 
what ill think ill do is this.....

master each song individualy first. Cause all of the songs will need different EQ and compression settings for sure. While i limit each song, ill keep a reference as far as the songs RMS levels of volume for that song, and use it as a reference point for the rest of the songs to be mastered.

once all the songs are individually mastered, then ill open a CD montage with all the songs in order, and apply fades as neccesary. Maybe even put a Waves L2/L3 in the master buss, and set it so its barely doing anything, but have it there just in case if needed be.

sound like a decent plan? or any other pointers i should be looking at?
 
actually....no. when i master each song individually, ill use Mbit +ultra from Izotopes ozone plugin pack.

that thing sounds pretty decent for dithering......

when i have the L2 on the master buss in the overall CD montage and arranging all the songs in a full CD, i wont be using dither of any sorts. Ill just have the L2 there for reassuring myself that its there .....if needed...........
 
bkkornaker said:
While i limit each song, ill keep a reference as far as the songs RMS levels of volume for that song, and use it as a reference point for the rest of the songs to be mastered.
Click to expand...
Not to put too fine of a point on it, but I personally wouldn't recommend using numeric RMS measurements for the reference. Identical RMS measurements on two songs of different sonic density will have different perceived volumes. Go by what sounds right and ignore the numbers.
masteringhouse said:
If one had started with the softer song and brought that up to the level that you feel it should actually be at, then tried to bring up the brick so that it's relative level was appropriate in comparison to say a softer ballad, you're screwed
Click to expand...
Hmmm...it's been my experience that the bricks usually need to come down anyway, not go up. Maybe that's because I can't/don't push as hard. Or just haven't been at it enough. Or don't do the metal thang.
continued said:
I guess what I'm saying in a nutshell is first set your parameters on the things that you can't change.
Click to expand...
That makes sense.

G.
 
bkkornaker said:
actually....no. when i master each song individually, ill use Mbit +ultra from Izotopes ozone plugin pack.

that thing sounds pretty decent for dithering......

when i have the L2 on the master buss in the overall CD montage and arranging all the songs in a full CD, i wont be using dither of any sorts. Ill just have the L2 there for reassuring myself that its there .....if needed...........
Click to expand...
Don't dither until the absolute very last possible step. If you go in and adjust fades (etc., etc.) again, you're going to have to dither again. Just use Ozone's limiter and dithering instead of the L2 on the final assembly.

Or if you're really set on the L2, put Ozone after it for dithering.
 
bkkornaker said:
master each song individualy first. Cause all of the songs will need different EQ and compression settings for sure. While i limit each song, ill keep a reference as far as the songs RMS levels of volume for that song, and use it as a reference point for the rest of the songs to be mastered.

once all the songs are individually mastered, then ill open a CD montage with all the songs in order, and apply fades as neccesary. Maybe even put a Waves L2/L3 in the master buss, and set it so its barely doing anything, but have it there just in case if needed be.

sound like a decent plan? or any other pointers i should be looking at?
Click to expand...

I think it's a little dangerous to work in terms of "individually mastered" and then "all songs in order". You've already done "individual" when you mixed the song. Mastering is when you put it all together. So in my opinion don't even start mastering the individual songs until you have a clear idea of the song order in the album and how you want it to flow.

Then, you need to keep the big picture in mind when mastering each song, since this is where it all comes together and you smooth out levels, unwanted tonal changes between songs, and transitions among other things. All decisions have to be made in the context of the whole. Otherwise, what you are doing is another mix of the songs, not mastering.

So in the mastering process you will do a lot of reference listening back and forth between tracks as you work on each individual song. Both during the process of mastering and then as you finish them, listening to them in their entirety in the context of each of the other tracks you have already finished. This will be an extremely time consuming process if you are not an experienced mastering engineer, which you aren't. But if you do it right you will learn a great deal from the experience.

I highly recommend you buy Bob Katz's book on mastering and study it closely. It is a wonderful reference book on this whole topic.

I agree with Southside that there's not much point in keeping RMS numbers. It's not a paint by the numbers kind of thing. You've got to listen and go by what your ears are telling you.
 
ok, this all makes sense.........but it now brings me back to my main question....

"master each song individualy, and the put them in a CD montage and arrange them by order and burn it..........

or.....

put all mixed songs in a CD montage, and then mastering effects in the main buss of the CD montage and master all the songs on the CD as a whole...thus the mastering effects would be the same setting for each song..."
 
or maybe this......

semi-master each song individualy (using EQ on the song as a whole, some compression if needed on certain songs) and polish the songs up, but not use any limiter or dither.....

then.......put all semi-mastered songs into a CD montage, and on the main buss have the limiter/dither, and that being the only effect on the master/main buss.
 
i guess where im really confused about is........i have the tools to "at home mastering".......and i know how to use the effects to master the songs.........

but i dont know the typical proper set up in which to lay these songs in a single audio editing program (such as Wavelabs 5) and get it done.

It makes perfect sense to lay all the songs in order, and master them together as a whole album..................................so would i apply mastering effects in the master main buss, and have all the songs all have the same effect settings?

or should i lay them all out in order (song order) and insert mastering effects in each individual song, and nothing in the master buss?

If anyone has used Wavelabs.......what is the proper way to actually set it up and actually start doing it? Open a CD montage (all songs in order)? or open each song (open wave) individually..........

sorry for being a pest about this.......
 
never mind guys.......ive got a plan of attack, and got it figured out.

thanks for all the help on this matter.........
 
good luck, i'm not too familliar with the overall componenets of mastering a full project, but me personally if i'mma master tracks 2gether i'mma wanna make sure they are all around the same overall volume with at least 3dB's of space left in case i'd need to apply any extra effects before evening out the levels..... well thats pending everything is mixed to a decent quality to start with........

but individually i'd lace the track til it's at least psuedo-mastered and then go for the kill once they all sound good with the designated 3dB's (or whatever your preference) left to work with....

i used 3dB's because it's enough for me to work with without overdoing it....
 
Each song needs to be individually mastered, but within the context of the whole. It's not an either or situation, it's a "both" situation.
 
You must log in or register to reply here.
Back
Top