Levels Before Mastering

E

Espadrilles

New member
Hi!

We're trying to mix our band's first demo (that we recorded on pc) and we were thinking on going into a studio for the mastering. My question is where should my mix levels be before all the mastering process? When i'm looking at my .wav file in Wavelab, should the levels be over, near or under the 50% mark when I normalise everything to 0 dB?

Thanks!!!
 
I'm not sure why you would want to normalize at all if you are going to give the mix to a mastering engineer. If you think that the mixes are so soft that you need to normalize just to get a decent level, it would be much better to remix them at a hotter level, rather than mixing them real soft and then normalizing.
 
If your mix is 50% below 0db, its WAY too quiet. Your mix should be somewhere between -6 and -10 db. You don't want to push it past 0 but you don't want it so quiet its hard to get the levels up. Do not normalize. I've heard it degrades the sound. Basically, you want to give the mastering engineer the best possible mix you can give him--as close to perfect as you can get it.
 
fenix said:
your mix should peak at 0db without normalizing.
Click to expand...
I disagree completely.... your mix should peak at the 0 VU level on your mixer, or whatvere 0VU is calibrated for on the mixdown unit....

This means you are gain staging properly, ensuring enough headroom for peaks, minimizing noise and distortion, and also giving the mastering engineer (whether it's yourself or someone else) enough room to work with....

If you blow your wad all at once on the meters at mixdown, there won't be any headroom left for the mastering stage.... especially if some EQ is required......!
 
Blue Bear Sound said:
If you blow your wad all at once on the meters at mixdown, there won't be any headroom left for the mastering stage.... especially if some EQ is required......!
Click to expand...

Assuming the mastering is done digitally. If it will be mastered with analog gear than a hot mix is fine.
 
I'm not sure I understand Blue Bear's point, (I got lost somewhere between 0dB and OVU) so I don't know whether I'm agreeing or disagreeing.

In my experience it's not that much of a problem if a mix peaks at 0 (whichever zero we're talking about), because the mastering studio will invariably create some headroom in the compression stage to add EQ or whatever is necessary.

But then again, this may have absolutely nothing to do with anything anyone is talking about...:confused: (it's okay... just ignore me).
 
I'm not sure if I understand well... When I do my mixdown, I put the master level at the highest possible value (but leave little room to avoid clipping). Even though I do that, my mixdown is still very quiet (below the 50% mark in Wavelab). Is that normal? Should I put more compression on my tracks in order to have smaller peaks?
 
...just forgot, how can I know that my mix is between -6 and -10db??? (it might be a stupid question but I still don't know...)
 
littledog said:
I'm not sure I understand Blue Bear's point, (I got lost somewhere between 0dB and OVU) so I don't know whether I'm agreeing or disagreeing.

In my experience it's not that much of a problem if a mix peaks at 0 (whichever zero we're talking about), because the mastering studio will invariably create some headroom in the compression stage to add EQ or whatever is necessary.

But then again, this may have absolutely nothing to do with anything anyone is talking about...:confused: (it's okay... just ignore me).
Click to expand...
0 VU is not 0 DbFS.... 0 VU is the nominal level for the mixer to put out its rated output.... which is general lower than than 0 DbFS...

My only point was that several db should be left for processing during the mastering phase, assuming digital processing (thanks Tex!)

Yes... I know the mastering engineer can easily drop the level, but it's also a good idea to avoid unnecessary gain additions/subtractions, which, in DSP terms is a multiplier of the orignal signal... Every digital manipulation that's gets done incurs additional round-off error that slightly diminishes the signal each time.......
 
1. I think the analogue at X and digital at X- is a bit irrelevant, as I presume the product would not be delivered to a mastering house on a reel. However, if it is, there is not really a standard rule, as headroom and tape compression depend on choice. Just make sure that the relevant information (especially machine calibration) is passed on to the mastering house.

2. If a mix is delivered for mastering on a digital format, get it close to 0 dB, but MAKE SURE IT NEVER CLIPS OUT.

3. You do not have to leave a couple dB headroom for a mastering house. Your material will in the vast majority of cases be transferred into another system (or onto analogue) for mastering, at which point the mastering engineer will have full control over the levels.
Note that in many DAW's the L/R bus quality (bitrate) will go down if the level drops, which is a good reason to keep the levels up there.

4. Refrain from "getting it close to 0dB" by means of applying limiters, compressors and all those other cute tools you might have. By all means use some light compression if you feel it benefits the sound, but only for that reason, and not to boost volume. Rest assure that any decent mastering house will have compressors and limiters many times the quality of the ones you might have.

5. If you HAVE to, use some very minor peak limiting (set for instance at -0.1 to -0.3 dB, not more), to avoid clips. However, the time to address such peaking is in the mix.

6. If you cannot attent the mastering yourself, make sure you give good instructions to the mastering house / engineer.

Finally, to master a well recorded and mixed track and make it sound brilliant does not have to take a long time.
To master a badly recorded and mixed track so it sounds reasonable can take forever.
 
littledog said:
Now sjoko's post i understood (and agreed with) completely!:p
Click to expand...

I understand it to just providing a mix that is the best you can do, properly recorded, and no fancy games that might resemble 'mix enhancing', as you'd in almost all cases just be stepping on the toes of the mastering engineer. That would include keeping the levels as 'hot' as possible (any format) without clipping (of course).

adding to the questions of mastering....how often would a mastering engineer see a mix that hasn't had a final fade out (or in) applied to the mix yet? Providing you would be present at the session (or provide great instructions...or a sample maybe?), would that feature be best left for this stage also? I sometimes think that fade outs (especially) can really make or break how a song might end. As a for-instance; ending a song that has a repetitve 4 bar phrase in the fadeout section...ending in the middle of the 3rd bar might sound ackward.
 
Following sjoko's advice is always a good idea. If your soundcard supports 24-bit audio, you might work at 24-bits in cubase and render your mix at 24-bits. Best to give the mastering engineer the best quality audio you can.
 
In my experience with mastering engineers I have been told do not let the mix surpass -2db that way it gives them 2db before it reaches 0 so that they can add the eq
 
You must log in or register to reply here.
Back
Top