Mixing Level Peaks

[Methlab]

When you mix, what do you like to get your peaks to before you master or throw an L2 on it? I was usually getting mixes that peaked at around -4 db, and then adding 5.5 db in the l2, but I'm wondering if tehre are btter ways..

[Bob Olhsson]

I try to mix with peaks between -10 and -6. Nothing is gained by pulling it up higher if you've got 24 bits to work with.

[chriscavell]

but I'm wondering if tehre are btter ways..

Yeah, don't use the L2 at all...make sure the levels are less than 0 dBfs...then send it to Bob!

[Mike Tholen]

NEVER mix up to 0dBfs!!!
If you have observed 0dBVU while tracking, your combined 2mix shouldn't come anywhere near 0dBfs...if it is, you have fuct your gain structure somewhere in your mix enviroment.
Gain structure is sooooo damn important and is even more important if you're mixing "in the box".
Proper gain structure starts in the tracking stage and if you **** up there it's impossible to fix later w/out screwing up the integrity of your tones.
When I have to mix "in the box" my 2buss peaks around -10dBfs which is still damn hot.
You will also be giving the mastering engineer the proper headroom to do the job correctly.fuuck

[chriscavell]

If you have observed 0dBVU while tracking, your combined 2mix shouldn't come anywhere near 0dBfs...

Not necessarily true. You can easily have rms levels hitting 0 on the VU meter with a dynamic range of certain attacks easily clipping digitally. There are many scenarios where you'd have to have your levels on a VU meter way lower than 0 to keep from digital clipping if mixing entirely itb. AAMOF, if you're hitting zero dBfs and 0 VU regularly in a single mix, chances are it's overcompressed....depending of course on the calibration specs.

if it is, you have fuct your gain structure somewhere in your mix enviroment.

That's funny.

Gain structure is sooooo damn important and is even more important if you're mixing "in the box".

It's just as important out of the box...I never implied gain structure wasn't important. Don't forget, I said "less than 0 dBfs".

Proper gain structure starts in the tracking stage and if you **** up there it's impossible to fix later w/out screwing up the integrity of your tones.

If you **** up that early, haven't you already screwed up the integrity of your tones?

When I have to mix "in the box" my 2buss peaks around -10dBfs which is still damn hot.

That's cool...-10 is definitely less than 0 and certainly isn't too quiet by any means. Perhaps I should have clarified, by less than 0, I simply meant DO NOT CLIP THE DIGITAL 2BUSS.

You will also be giving the mastering engineer the proper headroom to do the job correctly.

If the mastering is done ITB as well...it really doesn't matter so long as it isn't clipped. Regaining headroom is as easy as pulling a fader down...and the negligable one bit of resolution/dynamic range is not lost to the digital mix engine this way...but that single bit is usually a very minute point to make in such an argument. But, if everything digital to analog (and vice versa) is calibrated appropriately...headroom really shouldn't be an issue, but can creep up from time to time and require additional work to fix.

Bouncing a great deal less than 0 dBfs is really just a sort of "safety net" as far as I'm concerned if receiving a digital file to master...a good rule of thumb. But in reality and practice, it makes no difference to me if the peak is at -0.1 dBfs or -10 dBfs...so long as it isn't digitally clipped in the file that I receive.

Cheers mate, I honestly didn't think that saying what in essence amount to "don't clip the digital buss" would ilicit a fuuck response from anyone on this board.

-Chris

[Methlab]

thanks, this is great info..yeah i could send my stuff out to be mastered, but I get hired to do singer demos that need to be "loud" from me..so no outside mastering involved...I noticed the quieter my mixes, the better they sounded even when just run through the L2...thanks again for all the info and keep it coming.