Clipping master fader

[stevegalante]

I know it has been discussed a lot before, but anyways... I'm on PT HD Native w/ PT 10 HD (not tested this on 11 HD yet).
All my mix's tracks and auxes go to a mix bus (aux stereo input set to 0), where I have a few master plugs, and this bus's level is not clipping.
Then I route this bus to a stereo track (I usually don't BTD), and the level on this stereo track is exactly the same (obviously) as the mix bus, so again no clipping.
Of course I have nothing on the master nor the audio track.
BUT.. my master fader is clipping.
Why ???
Thanks

[mesaone]

The internal mixer is floating point in Pro Tools 11, so there's no hard clipping. The internal mixer in Pro Tools 10 is 48 bit fixed point (I think), which has a much higher dynamic range, so it rarely hard clips. Once it reaches the master, then you have the 24 bit hard clipping going on - and you can always clip the D/A that way, no matter what your internal levels look like.

I've encountered the problem where the internal (yellow) clip lights don't show up in Pro Tools 11. I don't know if that's what you're referring to.

[Raoul23]

Steve think of it like this (I maybe wrong) if all your mix tracks and auxes are set to 0db this will all end up adding up to more than 0db, like a motorway all lanes might be running ok but you come to a single lane and it gets congested. So all my sub mixes which feed to the master bus are turned down to around -10 hoping this will prevent clipping. Again I maybe wrong but this is what I do :)

I don't understand all the floating point 48 bit versus fixed 32 bit etc

[stevegalante]

Originally Posted by mesaone

The internal mixer is floating point in Pro Tools 11, so there's no hard clipping. The internal mixer in Pro Tools 10 is 48 bit fixed point (I think), which has a much higher dynamic range, so it rarely hard clips. Once it reaches the master, then you have the 24 bit hard clipping going on - and you can always clip the D/A that way, no matter what your internal levels look like.

That makes sense... in other words if my final bus and/or my final track are not clipping that does not mean that I don't need more than 24 bits, which is what the clipping on the master is indicating... right ? And if so, what's the use of 32/48 bits if we gonna make it fit into 24 anyways ? Are there any Avid "white papers" explaining this in detail (and the differences between 10 and 11) ?

[Darryl Ramm]

Originally Posted by stevegalante

That makes sense... in other words if my final bus and/or my final track are not clipping that does not mean that I don't need more than 24 bits, which is what the clipping on the master is indicating... right ? And if so, what's the use of 32/48 bits if we gonna make it fit into 24 anyways ? Are there any Avid "white papers" explaining this in detail (and the differences between 10 and 11) ?

The use of 32 bit floating point files is you can write out a file with huge dynamic range, exchange content etc. without worrying about clipping. Besides a slight hit on file space and disk bandwidth or wanting to exchange files with older Pro Tools versions or other software there is really no reason not to use 32 bit files.

The use of 32 (and 64 bit) internal floating point processing is that numerical steps along the way have huge overhead and you don't need to worry about clipping effects there. you have no choice about that, it's how Pro Tools 11 always works.

In principle all you need to worry about is pulling down the faders at the end so you don't clip the physical DAC on output. Now everybody should still cringe if somebody is being hamfisted/stupid and has the levels way out of whack.

No white papers that I am aware of, but none of this is really that confusing.

[mesaone]

Originally Posted by Darryl Ramm

In principle all you need to worry about is pulling down the faders at the end so you don't clip the physical DAC on output.

That's the big takeaway here. And, by itself, the fact that this is how things work is pretty amazing.

I've played around and pushed channels several hundred decibels above zero (with my monitoring system off, mind you). Bouncing from physical output at that point, with several Blue Cat Gain plug-ins in series on the master fader, each trimming back 60 dB, still got a clip-free print.

[stevegalante]

Originally Posted by Darryl Ramm

In principle all you need to worry about is pulling down the faders at the end so you don't clip the physical DAC on output.

So going back to my example on top of this thread what's clipping is just the audio output going to the DAC, not the file I'm recording (which is in fact showing no clipping). Why are you saying "In principle" ? Is it just a theoretical thing or is it just that simple ?
And what happens if I use 32bit files and need to import them in a 24bit session ?

[stevegalante]

An update on this, now from PT 12 HD...
my sessions are 24bit and my final mix aux buses (drums, guitars, keyboards, etc.) don't clip. Then I send everything to a "mix" aux where I apply some final stereo bus treatments (like compression, etc.) and it still does not clip. Then I send this aux "mix" bus to a stereo "print" audio track (I don't normally BTD) and this track still does not clip.
BUT, my master fader clips.
Also my Avocet monitor controller goes red...
Now, my questions :
- I guess the master fader clipping means it's going over my 24 bit resolution and it's clipping my DAC... so if I just lower it and the clips goes away, am I good ?
- what if I don't add any master fader ? Where does the 24bit truncation occurs ?
- why my final stereo audio track is not clipping if I'm recording at 24 bits ?

Thanks !!!

[JFreak]

There is always a master fader for every output, whether visible or not. Sometimes I think that PT should have a "show all master faders" menu command to make it clear for everyone; that whatever the output (physical or buss) there is always a master fader for it.

[JFreak]

Do you use pre or post fader metering? Switch to pre and you see the truth.