Mixing for the Mastering Engineer

[muxlow]

While mixing some songs, I slapped a little mastering plug on the master bus to boost the volume. When I killed the plug last night, I noticed that the relationships between various instruments were quite different. Guitars were too quiet, the kick was too loud, etc. I understand how the plug's compression was changing those relationships, but my question is...

1) Is it normal for the mastering process to change the relative volumes of different instruments or was that my own lack of skill?

2) If that's normal, how do you compensate your mix for some future ME's magic dust? Are there rules of thumb, or do you just wing it and remix to compensate after you hear a track?


Any input is appreciated!

[Kiwiburger]

My current thinking for a while now is to mix with absolutely nothing on the master bus at all. Not an eq, compressor, limiter, nor dithering plug. Nothing.

My reason for this is:

A - no eq, because you should be eq'ing each track which is far more effective
B - no compressor, because you should be compressing only the tracks that need it
C - no limiter, because your mix should not even be thinking about peaking anywhere near 0dBFS
D - no dithering plug - because you are going to export your mix at the same resolution as your project. Don't resample - and certainly not with a crappy DAW SRC.

But my main reason is simply - the ME has far better quality tools than I have. I don't want him having to use those tools in surgical repair mode, to undo anything i've done with cheaper tools. He only has two channels - give him complete access to those two channels.

I try to make my mix sound as good as I can, so it hopefully won't need much work in mastering. I use spectrum analysis, but never eq the 2bus.

My 2c.

[muxlow]

Originally Posted by Kiwiburger

My current thinking for a while now is to mix with absolutely nothing on the master bus at all. Not an eq, compressor, limiter, nor dithering plug. Nothing.

My reason for this is:

A - no eq, because you should be eq'ing each track which is far more effective
B - no compressor, because you should be compressing only the tracks that need it
C - no limiter, because your mix should not even be thinking about peaking anywhere near 0dBFS
D - no dithering plug - because you are going to export your mix at the same resolution as your project. Don't resample - and certainly not with a crappy DAW SRC.

But my main reason is simply - the ME has far better quality tools than I have. I don't want him having to use those tools in surgical repair mode, to undo anything i've done with cheaper tools. He only has two channels - give him complete access to those two channels.

I try to make my mix sound as good as I can, so it hopefully won't need much work in mastering. I use spectrum analysis, but never eq the 2bus.

My 2c.

All good advice, but you avoided my questions

Also, I should have said that I was ONLY putting a plug on the master so I wouldn't have to explain to bandmates why the rough mixes are "too quiet, can't it be louder?" and blah blah blah. All that would be taken off prior to sending stuff to the ME....

[Push845]

I say a great Compressor (SSL or Manley as a start) ONLY on the mix buss. It will tighten every thing up just enough.

[moogus]

If you need to boost the volume for a rough mix so the band dont complain (which, in a perfect world, you shouldnt but thats another story) but dont want to alter the mix by using a compressor, just turn up the master fader until your mix is clipped and loud enough for the band. This is pretty much what a mastering engineer does to make it loud as it is the most transperant way.



M

[planet red]

To me a 'perfect' mix is quite right once its slammed in mastering. Guitars and bass come up and attacks on kick and snare get softened. Vocals sit more in the mix... ect...

I check my mix thru a limiter constantly to make sure its going to be cool after mastering.

Ive been way more pleased in my masters when I'm mixing knowing its going to come back a brick and checking it in that form.

[Geert van den Berg]

Originally Posted by Kiwiburger

A - no eq, because you should be eq'ing each track which is far more effective

It completely depends on the material, if all tracks lack some top end, then putting on an EQ on the master or a subgroup is much more effective.

If you'd want to add some sheen to a drumgroup, this works much better when doing this on a subgroup than putting on highs on all the individual channels IMHO, because there are so many phase relationships on those multi-mic'ed sources, then you'd end with a phasey mess IMHO.

[u b k]

Originally Posted by Kiwiburger

A - no eq, because you should be eq'ing each track which is far more effective
B - no compressor, because you should be compressing only the tracks that need it

i assume those approaches work for you, so be it. i got an issue with the 'you should' parts, because they smell suspiciously like rules.

eq'ing groups of tracks is very different than eq'ing individual tracks. if you have a pultec pushing a little air on everything, you'd probably be reaching for less hi eq on all the individual tracks. less phase smear, more pultec goodness.

as for mix compression, one of the main reasons for slapping a comp on the 2bus is not to tame individual instruments, but rather to get everything interacting with each other in a way they don't without the comp. mixing into the comp is a whole different ballgame than squeezing single elements and busses.


gregoire
.
.. ubk
.

[muxlow]

Thanks for all the advice / information!!


The general consensus seems to be:

1) Yes, mastering is going to change the relationships between instruments IN ADDITION to making everything louder

2) Yes, you have to compensate for those anticipated changes while mixing

3) Check your mix through a limiter / compressor while working so you have an idea of what it's going to do when the ME starts squishing it

4) Get a test track from the ME before he does the whole album and be prepared to make any necessary adjustments


Does that about sum it up?

[bobdemaa]

I do a lot of budget mastering, and mixing. there's a few suggestions I'd like to make.

First talk to your mastering engineer. I would absolutely love it if the cllients would talk to me before making their mixes. You can address concerns you have right out of the gate and establish a plan. In some cases you might want to supply stems if you're concerned about bass levels and such. other times, the engineer might be able to identify a better signal path, or at least become aware of what your path is. It can make such a huge difference in what the mastering engineer can accomplish.

Second. what kind of 2-buss compression are we talking about and how is it being used. I like to mix with the Alan Smart C2 on my master from start to finish. This way I'm putting things in the mix where they really are, and I'm getting the punch and snap I want to hear. BUT, I'm not overdoing it either. Ratio is at 1.5, The meters are moving for the kick and snare, and when everything hits at the songs peak. but It's not squashed. If you're just throwing an L2 or L3 plug-in on the master fader, then I say don't mix this way, and don't send something with multiband compression or limiting to the mastering. The More Dynamic and open sounding the master is generally, then the better job the mastering engineer should be able to do.

When it comes time for band dubs or A&R dubs, then it's not uncommon to do something to "pump it up" for their listening, but keep a distinct difference in mind of knowing what is the right mix, and what your plan is with the mastering engineer, versus the dubs for the band. Either that or you could try just letting them in on the procedure. Give them exactly what you're sending the mastering engineer and tell them how to turn up the volume knob, but I know how this argument goes...

To answer specifically.

1) No, It depends on the compression being used. Some are very transparent and will effect the balance of the freq's or instruments differently. It might not be lack of skills as much as lack of tools.

Hope this is helpful.

[NathanEldred]

I say mix with anything ON that is going to change the end result. It doesn't mean you have to print it that way, but it's going to make yours and your ME's life easier, because there will be less surprises in the end. It's really common sense IMHO.

[edvdr76]

I try to get the mix to sound as good as I can and at the end I run everything through a Summit DCL200 tube compressor to kind of glue everything together. I go about 3:1 on the Summit. I set the gain around 5 or 6. I just mainly use it for the sound. It makes everything sound wider. FAT AND WARM! I don't compress too much, I let the mastering guy do the heavier compression.

P.S. If youre just trying to make a reference cd louder to keep the band exited, then what I do is listen to the playback from your cd burner, bring the gain up all the way to the red, right before you hear it clipping. I might be wrong but that's what I do to keep'em happy!

[bolehnggak]

Hi,

Is it okay if I continue the thread a little?

I have a question though. I have no budget for mastering, but I have Waves L3plugins, and I tried to mix so it would be very near to the quality of mastered mix. So right now I always put the L3 multimaximiser in the master bus, with the CD mastering setting.
For those familiar with the Waves plugin, it's threshold is -5db, with the output gain of -0.2.
But I don't know how hot should I mix. Should I get many gain reduction, or just peak a little. Because most commercial records that I often refer to has average gain of -12 to -14db RMS, and I can only achieve that loudness by mixing very hot, but the sound isn't satisfying, but if I got the sound of the mix that I wanted, it would be too low. But honestly I want it to sound as loud as those records. (Of course, for the suitable genre, I won't mix a soft jazz music as loud as a guitar driven rock band)

Any suggestion?
Thanks in advance.

Ari

[Bob Olhsson]

Beware that a fragile mix balance that changes significantly from a musical standpoint when compressed is likely to have some pretty big problems on the air when it gets compressed a lot more.

I used to use a compressor as a test adjusting the balance so the mix "works" no matter how much it gets mashed after it leaves my hands. Then I'd leave it off.

[hourglass]

Originally Posted by bolehnggak

I have a question though. I have no budget for mastering, but I have Waves L3plugins

Stop giving your money to Waves and you'll be able to pay a mastering engineer to do it right.

ryan

[joaquin]

Originally Posted by Kiwiburger

D - no dithering plug - because you are going to export your mix at the same resolution as your project.

I do not know about 24bit fixed, but if you are working with 32bit floating, it's recommended that you dither even though you are not really going for a smaller resolution bounce; ...for the internal mixing one is higher (32bit floating).
Perhaps someone could confirm this.