M_Glenn_M


As I understand now, it's more a function of leaving dynamics (minimal compression and limiting)  than levels.
One could over comp and limit on the tracks and do the same damage.
Right?
So it seems we need some guide that indicates acceptable dynamics rather than levels at master.

Yes.

If you export your mix in 32bit floating point, the mastering engineer can harmlessly adjust the levels by whatever amount is necessary and appropriate before the start of processing and still achieve whatever level is desired at the output.

But if you compress/limit things too much, there's no good way to undo that in mastering. 

Lanceindastudio


DON'T normalize before mastering. Give them 3-6db of room to work with. DON'T limit or compress on the master buss. No need to discuss  further. End of discussion. 


Lance

As I asked in my post. Why? Why can't they just turn it down 3-6 db if they need additional headroom.  I only have one peak hitting the -0.1 ceiling and the song is not clipping anywhere but that one small peak. What is the scientific/sonic difference.  (If the master buss is clean)
 

Chregg


"*I don't understand why anyone in the modern world would use anything less than a 32bit fp delivered to mastering. " to be fair most people probably do Drew

Really? Well, like I said, I don't understand why. 

But even with 24bit, the only really important difference is you just need to make absolutely sure you aren't clipping when you export...

I'm also not getting the reference to 24 & 32 bit export?
Also, it's easy not clipping on the Master buss, but to do it and leave some dynamics for the engineer makes sense.
I would guess a look at the waveform after export would be a clue?
I guess if it looks like a brick, it's over done?
If the average mass of waves (what would you would call that? The mean levels?) is low with lots of spiky bits, it may benefit some gentle compression?
So (after HP and LP filtering) only use compression and/or levelling until the average or mean level is "-X" DB and yet not redlining would be the best? 
Anyone have a "good" waveform to judge by? or is that genre specific?
Two examples I have:
https://dl.dropboxusercon...4/good%20and%20bad.jpg

I think there is much misinformation about mixing an mastering. Don't handicap your mixes by not using your tools. Mixing should be to your ideal conclusion. I use 2bus compression to varying degrees on pretty much everything I do. When I'm done it pretty much sounds like a record that you have to turn up a bit more. The mastered version should sound like what you sent in just a little louder and maybe some more tonal unity across the collection.

I have a few guidelines that I follow that are based on feedback I've gotten from MEs.. No overs, ever but I will go to 0dbfs (but I don't shoot for it). I will manually duck any standout transients that keep the me from pushing the overall program a bit louder with makeup gain on the 2bus comp. All the mastering engineers I use have analog chains so as long as I don't clip their DA then it's not a big deal for them to do whatever they need to do in the analog domain. 

Of course I leave several db of room for a mastering engineer if I have any controll of what is finally going to be sent to a ME. However clients sometimes want a loud(ish) mix, hence the normalizing with nothing on the master bus, but then will take that mix to a ME themselves at a later date. This seems like the only way I can deliver a semi loud mix with the very least amount of damage. Actually I don't see any difference in normalizing than using a brick limiter and only boosting the level until one little peak hits the ceiling. Isn't that the same thing more or less as normalizing?

bandso

Lanceindastudio


DON'T normalize before mastering. Give them 3-6db of room to work with. DON'T limit or compress on the master buss. No need to discuss  further. End of discussion. 


Lance

As I asked in my post. Why? Why can't they just turn it down 3-6 db if they need additional headroom.  I only have one peak hitting the -0.1 ceiling and the song is not clipping anywhere but that one small peak. What is the scientific/sonic difference.  (If the master buss is clean)
 

Aesthetics, attention to detail and professionalism - Its like me asking you to put the frosting on a piece of cake before you deliver the cake instead of handing me the tube of frosting to apply myself. It is nicer for me not to have to do an extra step. I like my mastering engineer as happy as possible when I turn in a mix.

Sound wise, if you are not limiting, peaking etc. you are not going to have any sound issues if they just turn down the main for you.
Normalizing - I just dont like to go there JUST IN CASE it might be messing with the mix.

Lance

I thought Normalizing just raised the wave until the peaks were at the Normalizing setting. IOW all the waves are raised.
I thought levelling simply pulls the greatest peaks down leaving the others untouched and thereby allowing the rest of the waves to be raised without peaking.

Normalizing boosts the entire clip/song louder until the single loudest peak reaches the normalize setting. As long as this is below 0 then there should be no clipping.

M_Glenn_M


I thought Normalizing just raised the wave until the peaks were at the Normalizing setting. IOW all the waves are raised.
I thought levelling simply pulls the greatest peaks down leaving the others untouched and thereby allowing the rest of the waves to be raised without peaking.

That's a much longer discussion. I highly recommend a book by a guy who calls himself "Mixerman", Zen and the Art of Mixing. This book will take you through a whole chapter on the subject, and I wholly agree with him on his stance. It's available via Kindle as well. Go check it out, because it's got some real-world applicable stories and experience. There's some profanity in there, but you'll understand why when you get to know the author. It's a good book, and short.