digital gain

[Farview]

Ok then... so what about mastering?

When I EQ/Limit/Compress my final mix in the mastering process, I apply gain to increase the overall volume of the track. I limit to -0.1 db as to avoid any digital distortion from encroaching the 0db mark. Essentially speaking, isn't adding gain to a final mix, the same as adding gain while in the mixing process?

If we left everything at -6db we'd have extremely low-volume CD's, which most people won't want to purchase.... right?

The difference is that mixing is mixing and mastering is mastering. The goals are different.
You will find that if you leave the headroom in the mix, when you go to master it, it will always sound better. When eq'ing, even subtractive eq'ing, you are adding gain. If your mix doesn't have any headroom, you will take the chance of clipping the eq and everything else down the chain.
Mixing and mastering are two different processes, don't try to get a finished product before you are finished.

 

[chrisharris]

This thread has me confused, and I don't think I'm all that thick, so I obviously don't understand something.

Vocals are typically the last thing I record. Now, none of my individual tracks (kick/snare/toms/oh's/ bass, guitars, etc) are clipping when SOLO'D with the master fader at 0db. However, with everything running, the meters go nuts, so I pull the master fader down...maybe it's down by as much as -11db by the time I track the vocals.

So now let's say I have a lead vocal track. In CEP/AA, the very first thing I do with that track when I start mixing is to turn the individual track up by 6db (fader adjustment) then I pull the volume envelope down by 6db. It's just easier to envelope the track UP as well as down since CEP/AA defaults the envelope to 0db. In my mind, the only gain adjustment to the track at this point is the -11db from the master fader, since the +6db and the -6db wash. Ok, maybe I've got a compressor in the chain which increases the output by some amount, then I've got reverb (reduces volume usually)...etc. The long and short of it is that by the time I've got all this crap on the vocal, I may have the individual fader for the vocal channel set to +15db to make it sit in the mix correctly.

Now that +15 vocal track will damned sure clip if the master fader is set back to 0db. Very audibly and nastily. It does not do this with the master fader set at -11db. Am I just deaf? I mean, I'm pretty sure I'd notice if I was just "turning down a clipped signal."

 

[Blue Bear Sound]

Apparently, what we have here is a failure to communicate!

 

[Massive Master]

I do something very simple - I don't track and mix much anymore, but when I do, NOTHING reaches over -6dBfs EVER. There are engineers that I've suggested this to also. Of the ones that have started doing this, the mixes I receive from them are SO much better sounding that even I don't get it.

I can only assume that it's a cross between a few things -

1. They're not clipping stuff anymore. Whether you "hear" it or not, (many light clips aren't that audible for the most part, but cause a lot of fatigue).

2. D-to-A reconstructive distortion* is basically taken out of the equation. IMHO, this should only happen at the mastering and/or consumer level. Actually, I wish there was no purpose for it at the mastering level either, but I'm not naive enough to think that people will ever go for "good sound" over "sheer volume" anymore. And if a compromise is going to happen, it should happen at that point, on converters that handle it in style, then of course at the consumer level, normally on crappy converters that I have no control over.

3. They stop concentrating on "bit hoarding" and start concentrating on making stuff sound good - Which is normally compromised in the effort to hoard bits.

Set a decent level that you KNOW isn't going to clip - then forget about it.

YMMV

 

[chrisharris]

Apparently, what we have here is a failure to communicate!

Okay...thanks for clearing that up, lol.

 

[chrisharris]

If none of the individual tracks are clipping, but show clipping when summed, then yes, you can lower the master level to avoid the clipping.

That I understood.

No. This is wrong. You cannot just turn down the master fader. If the master fader is clipping when set at 0db and you turn it down, you are turning down a clipped signal. The light will shut off, but it will still be clipped.

That I don't understand. Like I said, I can set the master at 0db and make an individual track audibly clip like a monster...doesn't happen when the master fader is pulled back.

Somebody asked about the processing chain. Here you go.

 

[Farview]

That I don't understand. Like I said, I can set the master at 0db and make an individual track audibly clip like a monster...doesn't happen when the master fader is pulled back.

When you hear the clipping, what is being clipped? If the d/a is being clipped then, yes, turning down the master fader will help. If you are clipping the mix buss, it will not. If you are doing both at the same time, pulling down the master fader will stop the d/a from distorting, so you will hear an improvement, but you could still be clipping the mix buss.
The picture of the signal chain still doesn't tell us where in the chain the meters are hooked into. It also doesn't tell us if it is a floating point or a fixed point mix buss. That would be the biggest factor in determining if the master fader will help you.

 

[chrisharris]

It also doesn't tell us if it is a floating point or a fixed point mix buss.

32 bit float. That seems to indicate we're dealing with a floating point somewhere, but hell if I know. I just know what my ears tell me, which is that a signal may distort when the master fader is at 0db and NOT distort or clip when it is lowered. Some of the analog old timers seem to have a hard time with this concept, but my complete ignorance aside, I'm pretty sure about what I hear. If I'm clipping something that is completely inaudible when a track is solo'd, then I guess I couldn't care less, lol.

 

[Farview]

32 bit float.

You should be OK. This is not the case with every software or device. If you get used to doing things this way, you may have to unlearn this bad habit in the future.
It's like when you are driving a car and you come to a stop sign. You take your foot off the gas and hit the brake. You could argue that you don't have to take your foot off the gas, you can just use the other foot to hit the brake. The car will still stop, so what difference does it make?
Sometimes the only reward for doing things the right way is knowing that you are doing things the right way.

 

[chrisharris]

cool, thanks. This has been an educational thread, mostly just educating me about how ignorant I really am, lol.

 

[dobro]

Okay, I said I'd report back, so here it is: after checking on another couple of boards, my conclusion is that because Cool Edit and Audition use floating-point calculations, there won't be any distortion if you raise the master volume above 0 dB. (Don't ask me to explain it - I'm a dumbass about this stuff - I'm drawing on the opinions and knowledge of others here.)

I would also like to report my pleasure at seeing a very useful thread unfold which exchanged ideas and information yet which didn't exchange comments about the intelligence or otherwise of the persons doing the exchanging. Always a relief in my case.

 

[BRIEFCASEMANX]

this thread has been educational. Thanks everyone.

 

[Change of POETS]

I'm with Chris on this one I guess... inevitably I always tend to rely on what my ears tell me, and when I drop the master fader in a mix, they tell me there is no distortion, and I've always been "ok" with that...lol..

But I will now attempt my next few sessions with mixing at a peak of -6db as Jason suggested, and see if it does in fact improve the mix and master afterword.

Great thread guys.