How to handle projects with tracks that were recorded too loud?

  • Thread starter Thread starter peritus
  • Start date Start date
Status
Not open for further replies.
peritus

peritus

The not fountain head
Ideas?

Based on all the stuff I've read here and elsewhere, our target for peaks should be between -14 dbFS and -12 dbFS... And, it's my understanding that the movie industry goes even lower (somewhere between -20 dbFS and -14 dbFS)....

Anyways, where I'm given a project to mix with peak levels approaching -3 dbFS or higher/worse, what should I do? Aren't the dynamics destroyed, for the most part, already?

Two guesses/options on this (I like second better than the first)(both are rendered and are not using the plug-ins as inserts):

1) Use a "Gain" plug-in (I'm in Pro-Tools), obtaining the peak values for each track and, if they exceed my desired dbFS, move the slider to the left until it lessens the dbFS to where I want it...

2) Use Waves L1 (or similiar mastering plug) and, using quantization but without using dithering or shaping, bring down the levels using 0.0 threshold and an approximation of the desired "ceiling".... Double check using the Gain plug-in's "scanning mechanism"...

:D I'm so obsessive...
 
how about using the normalising audiosuite, but set yer output max @, say, -14dBFS.


What this will do is search thru the audio file, find the single highest peak, and either pull it down or push it up to whatever you've set it as.


A bit like a tape input trim on a desk really...
 
peritus said:
Aren't the dynamics destroyed, for the most part, already?
Click to expand...

just curious as to why you say this...surely the dynamic range is still the same? i mean, just because they tracked hot doesn't mean the preamp will be doing some osrt of undercover compression, right?


or is there something fundamental i'm overlooking?
 
MessianicDreams said:
how about using the normalising audiosuite, but set yer output max @, say, -14dBFS.


What this will do is search thru the audio file, find the single highest peak, and either pull it down or push it up to whatever you've set it as.


A bit like a tape input trim on a desk really...
Click to expand...


Makes sense.. Good suggestion... I read somewhere to stay away from the normalizing plugs... I guess it prolly has the same innards as the gain plug though...
 
MessianicDreams said:
just curious as to why you say this...surely the dynamic range is still the same? i mean, just because they tracked hot doesn't mean the preamp will be doing some osrt of undercover compression, right?


or is there something fundamental i'm overlooking?
Click to expand...

I said that, because there are "hidden peaks", if you will, between samples...

And with the dbFS so high, they distort.. Thus, causing the upper dynamics to be hacked off.. (edit: or rather, the transients were not accurately captured at the time of recording)

What I failed to mention is that, when I lower the gain of my file, I may be making things worse.. I mean, if those "hidden peaks" are already gone and I'm hacking off the lower dynamic range.. I'm losing valuable information.. argh
 
Actually.. I must revise what I said...

It's the "nominal" level that should be in the ranges I mentioned.. dbVU would be a classic method of determining this... and, according the AES specification document I read,

0 dbVU / -14 dbFS / 85 dbSPL are all linked to one another... (edited)

http://www.aes.org/technical/documentDownloads.cfm?docID=65

Edit: I've decided to close this thread... Warning: It may contain misinformation, on my part.. Read the document immediately above...
 
Last edited:
Status
Not open for further replies.
Back
Top