mastering questions

Rusty K

New member
Hello,

Is it true then that if I were going to send a mix to a mastering house, I should leave the file as 24bit and do no dithering? Second leave at least -3db of headroom? I have heard that -6 to -10db for 24bit is even better.

As to minimizing the amount of compression applied, this is dependent upon the type of song if I'm understanding. I would think that one would want the final mix to be pretty close to the final product post mastering. If it's a "balls to the walls" rock n roll song that you want to run up close to 0db, how can you avoid over compressing in the mix? Could it be more a question of the trasparency of the compression? Like if I can easily hear it being stepped on in the mix then that won't translate well in the mastering process?

Am I in the ball park with these "rules of thumb"?

Thanks,
Rusty K
 
Hello,

Is it true then that if I were going to send a mix to a mastering house, I should leave the file as 24bit and do no dithering? Second leave at least -3db of headroom? I have heard that -6 to -10db for 24bit is even better.

As to minimizing the amount of compression applied, this is dependent upon the type of song if I'm understanding. I would think that one would want the final mix to be pretty close to the final product post mastering. If it's a "balls to the walls" rock n roll song that you want to run up close to 0db, how can you avoid over compressing in the mix? Could it be more a question of the trasparency of the compression? Like if I can easily hear it being stepped on in the mix then that won't translate well in the mastering process?

Am I in the ball park with these "rules of thumb"?

Thanks,
Rusty K

There's no reason to ever be up around 0db before mastering. How "balls to the wall" a song is has nothing to do with how loud the average level will be before mastering.
 
There's no reason to ever be up around 0db before mastering. How "balls to the wall" a song is has nothing to do with how loud the average level will be before mastering.

I thought it was clear that I was referring to post mastering. However, the song needs to be mixed like you want it to eventually sound otherwise how do you know what you've got? I don't see how you can leave all the compression/stomp to the master. Again I'm speaking about a hypothetical hard driving rock n roll or metal band sound.

I've never had anything mastered by someone else so I don't really know how the process works. Do you have to tell the mastering house/person what you want specifically for each cut? Or should they be good/experienced enough to process to song/style specific industry standards without my input?


Rusty K
 
However, the song needs to be mixed like you want it to eventually sound otherwise how do you know what you've got? I don't see how you can leave all the compression/stomp to the master.
If your mix doesn't have "that sound" before mastering, it won't have it afterwards either.

And "that sound" is not going to be gotten by mastering compression; all the mastering compression is going to do is make things louder and - if pushed as hard as it sounds like you might want it - lousier.

I'd talk to the ME first and ask them what they want. Most likely they are going to want to work with your raw mix and whatever levels it naturally comes out at. Having you push things halfway home is just going to make their job twice as hard. However, they may accept a second, reference mix (yours or someone elses) to use as an example of how you'd like to end up sounding.

G.
 
I thought it was clear that I was referring to post mastering. Rusty K
This is what threw me off.
I would think that one would want the final mix to be pretty close to the final product post mastering. If it's a "balls to the walls" rock n roll song that you want to run up close to 0db,
 
Last edited:
Your "stomp" and everything else should be in the performance and sound. If you're going to throw compression or anything else on something, it should be for a reason other than volume.


I think we're talking past each other. I'm more interested in the details of the mastering process so that I know how best to prepare my tracks for that stage of production. You offer a blanket statement with no roadmap and it seems contrary to what I've learned/heard from engineers who have told me "don't over compress at the recording stage. Leave me some room for compression at mixdown/master". If the "sound" is all at the performance stage (by your measure) then what is the need for mastering?
 
If the "sound" is all at the performance stage (by your measure) then what is the need for mastering?
Mastering adds a bit of fine sanding and polish, it is not meant to do the carving and shaping. Mastering is meant to prepare final mixes for sale, not create final mixes.

Everything is front-loaded. Start with a performance that doesn't have the energy you're looking for to start with, and you'll never get it in the engineering. Lose it in tracking, and you will never fully recover it in mixing. Lose it in mixing and you'll never fully recover it in mastering.

Just get you mix sounding right sonically and don't worry about the volume. If they have "that sound", they'll have "that sound" at -14dBFS RMS as well as they will at -10dBFS RMS. If they don't, then the problem is earlier in the production chain, and expcting the ME to pullthe sound out of his ass is not the way to go.

Send your tracks to the ME at whatever volume the mix naturally comes out at, making the mix sound as good as possible, and let him worry about final levels and do the volume work.

G.
 
SouthSIDE Glen,

Sorry I missed your post...thanks for the response.

This "mastering thing" is a mystery to me so I'm just trying to get a handle on it. Could I make the general statement...."mastering is to bring a mix up to industry standard for CD production and distribution"? Sort of a "normalization" for lack of a better term?

I open popular CD tracks in an edit window and see them smashed up to 0db. Of course "treatment" depends on the type and style of the song but I don't want a listener to have to adjust the volume for my hypothetical rock n roll tune after listening to a U2 or Green Day cut. If the last track was "intense"
then I want my song to be just as "intense".
 
SouthSIDE Glen,

I get what you are saying but compression does shape the sound. It can bring cohesion to a mix. If I'm to mix the sound I want then wouldn't any further compression to raise the level to industry standard in mastering threaten that original mix?

Are you an ME? Would you advise me not to over compress my original mix? If yes then how do I get the sound I'm looking for if it's sounding like it needs more compression or cohesion. Again I'm speaking about a hypothetical mix.
 
You offer a blanket statement with no roadmap and it seems contrary to what I've learned/heard from engineers who have told me "don't over compress at the recording stage. Leave me some room for compression at mixdown/master".

I offered a TRUE blanket statement because I'm not going to offer a road map when I, or you, don't know where you're going.

My statement is not contrary to what ME's have told you. Where did I say you need to over-compress, or compress at all, at the recording/mixing stage???? The problem is, like Glen so subtly pointed out, that you're expecting the mastering process to do what should already occur during the recording and mixing stage. You can have a song averaging -20db and still have this "energy" you're looking for. Mastering, in this conversation, simply brings the overall volume up.

Of course, mastering is much more than that. It involves arranging songs in the best order, it even involves how long the gaps are between songs, among many other things. But it's not about pulling out the "umph" that isn't already there in a mix.
 
I get what you are saying but compression does shape the sound. It can bring cohesion to a mix. If I'm to mix the sound I want then wouldn't any further compression to raise the level to industry standard in mastering threaten that original mix?
Yes, you are right that compression, EQ, and many of the other things that a good ME will do does change the sound. I didn't mean to imply that mastering is completly transparent. On the other side of that coin, the goal of a good mastering engineer is usually not to try and create the sound of the mix, but rather to make the mix that is there shine. Maybe a way to state it is that the mix down should bring you 90% of the way home, and mastering will add that last 10%. But don't expect that last 10% to give your mix a character that isn't already mostly there.
Are you an ME? Would you advise me not to over compress my original mix? If yes then how do I get the sound I'm looking for if it's sounding like it needs more compression or cohesion. Again I'm speaking about a hypothetical mix.
I am not a professional ME, 90% of the audio work I do is in mixing. But in that capacity as the creator of the 2mix, I have worked with MEs, and also have made friends of and learned much from many of the pro MEs who do hang out on this board. I'm sure they will check in here sooner or later with their take on all this, and I will defer to them.

But, in the meantime, based upon my experience, the more an ME has to work with, the happier they are. Put kind of bluntly, let them do their job and don't try and do it for them. In short, once the 2mix is done, let them take it from there.

Now, that doesn't mean that you shouldn't do what you can to make the mixdown sound as good as possible. Just the opposite; the better the quality of the mixdown, the better the job the ME will be able to do with it. But that means getting the mixing right. Getting the mixing right means getting the mixown right when you mix it down, not trying to "fix it up" after the fact. Get the tracks to work together by working on them at the track level.

If Instrument X needs compression to get the kind of punch you're looking for, then give it to that track, and then find the best way to fold that track into the mix. But to apply blanket compression to the entire mix because of one track that needs a specific kind of punchup is kind of like firebombing an entire city just to take out one command post :).

And remember, when you see U2 or Green Day squarewaving to get volume, there is often more than just compression going on. Often times they will incrementally add layers of light compression, tweaking and adjusting as they go, maybe with a *little* MBC to anticipate the next compression, or with some M/S-based EQing to pull out a track that got a bit buried by the last compression, etc. Then perhaps some hard limiting of just a few dB, and sometimes even some little bit of purposeful clipping through an additional stage of A/D conversion to give that final volume boost (and in the process, chopping off the peaks.)

I won't go into the pluses or minuses or other details of each step in that process, but on that pro level usually (there are always exceptions) the ME goes through all those steps of work not so much to make the mix sound greatly different, but simply to make it sound louder *without* sounding a whole lot different.

G.
 
Last edited:
Thanks for all your help and your time explaining this.

I don't really expect to compete with the big time pros with my little home studio and I still have much to learn about compression in general. I have produced some mixes that have stood up pretty well over time, at least to my ear, but I do know to take it to the pros if I want it done right. I just want to make sure I have the tracks in good condition for them to work with.

I do hope one of those ME's stops in. I'll be looking for them!

Thanks again,
Rusty K
 
Rusty, how about sending me a one minute or more sample of what you have. I may be able to show you what some mastering might get out of what you already have. I'd recommend sending just the raw mix. No processing after mixing. And be sure whatever reverb you have is NOT overdone. Willing to give it a try?
 
AtlantaCC1,

Sure..that would be great...but a couple of things. I just switched over to Pro Tools from Adobe Audition and I'm on my first mix right now. I'm still learning how to get around in PT. I've also been playing around with the Massey demo plugins including their mastering plug. Settings won't save with the demos so I have to set the mix all back up again every time I restart the session. Will you give me a day or so to get it to you? I'm close to a final mix on a simple little acoustic song, three guitar tracks and vox. No drums so it's just a soft ditty. I'll take the mastering plug off, refine the verb and mix it to say -10 db 24bit? Would one min. at that rate be too big?

Do you want me to send it to an email address?

Thanks
Rusty K
 
AtlantaCC1,

I've got a 40 sec 24bit segment at a level of between -6 and -7db peak. I hope I don't have too much verb on it. Just let me know what you want me to do with it.

Thanks,
Rusty K
 
Hello,

Is it true then that if I were going to send a mix to a mastering house, I should leave the file as 24bit and do no dithering? Second leave at least -3db of headroom? I have heard that -6 to -10db for 24bit is even better.

As to minimizing the amount of compression applied, this is dependent upon the type of song if I'm understanding. I would think that one would want the final mix to be pretty close to the final product post mastering. If it's a "balls to the walls" rock n roll song that you want to run up close to 0db, how can you avoid over compressing in the mix? Could it be more a question of the trasparency of the compression? Like if I can easily hear it being stepped on in the mix then that won't translate well in the mastering process?

Am I in the ball park with these "rules of thumb"?

Thanks,
Rusty K

The purpose of staying away from hitting 0 dBFS is to have a cushion for the ME to work with and ensure that no clipping is occuring. Scales measure what a "clip" is differently, usually it is considered 3 consecutive samples at 0. By having a peak measure -3 or less it helps to ensure that no clips occured in the mix. Intersample peaks (basically peak samples reconconstruted during D/A conversion) can measure even higher than 3 dbFS so some are even more convervative with their peak levels and shoot for -6 or more.

Personally I'm concerned more about crest factor (the headroom between the average level and peak). If a mix is overcompressed it can't be undone easily, while if it's "undercompressed", compression can always be added. If you are compressing a mix I feel it's better to be conservative since you don't have a good measure to compare the song with the rest of the album. You can then add as needed rather than trying to undo a destructive process.

OTOH, if you are a seasoned mixer, with good gear/ear, and absolutely know where you want the mix to be, go for it. You can also supply two mixes one processed, one not, and let your ME decide. If nothing else it's an indicator of what you want your mix to sound like.

Hope this helps ...
 
masteringhouse,

Thanks for your input. I guess my lack of understanding the mastering process makes it seem intimidating. Like.... you pass your mix through some big black doors and out pops "the master". I forget there is actually human interaction with the ME.

I have read other threads where the mastering results were described as not that good. As in any profession I'm sure there are good and there are not so good. I've also had engineers throw a mastering plug (which I have) on to the mix with noticable results. Because I am primarily a musician and only an intermediate engineer, I'm not yet confident of my mixes and so would be reluctant to spend the money on having a pro master it. I would however like to gain more confidence so that I could feel good about recording at home and sending the tracks on to an ME. That's why I'm here and I appreciate you time explaining.

Rusty K
 
Intersample peaks (basically peak samples reconconstruted during D/A conversion) can measure even higher than 3 dbFS so some are even more convervative with their peak levels and shoot for -6 or more.
...
Is the concern about the D/A conversion because you master with analog equipment?
 
Rusty, the email in your profile is returning to my email. You can click on my name in this post and send me an email with your correct one written in the body of it. We'll communicate that way. You may want to go to your Homerecording profile and re-enter your current one, though.
 
Is the concern about the D/A conversion because you master with analog equipment?

As far as input levels, yes. Obviously you can turn it down in your DAW before going to the converter, but if you can avoid any processing before the chain it's better by a small amount.

One also has to consider this in the final product regardless of analog processing, but since most folks would not be happy with an overall level that is 6 db down from the norm, we don't think about it for very long. :)
 
Back
Top