Final Mixdown Master Volume -6db?

lurgan liar

lurgan liar

Jimmy Page XXVIII
Hey guys, i have recently mixed a few songs so that the master volume peaks around -0.7 Db....

I have now realized that i shouldn't be peaking that high on the master volume ... because i need to leave headroom for mastering...

So i was wondering what would be the standard master volume level for the final mix ....

I have Izotope Ozone and PSP Vintage warmer to add to the stereo mixdown ....

Can any one offer any advice on how i could use these plug ins together as i really like the sound of the PSP Vintage Warmer...

Hope I have explained myself ok... :)
 
Well at this point I wouldn't turn it down just to add headroom . . . the ME can reduce gain just as easily as you can. So long as you haven't clipped anything, it should be fine.

However, if you look back at your chain of plugs and see that you've unnecessarily turned up the mix at some stage, you could reduce it there.
 
I would set it so your peaking at -3...that being said, I would MAYBE throw some ever-so-LIGHT limiting on it with the threshold at -3...don't leave it for the ME to bring down. some of those guys hate doing things they shouldn't have to.
 
That seems to be an odd theory to me . . . here's an example, let's say you have a mix with an EQ problem you haven't noticed, the ME fixes that upfront and *bam* headroom instantly appears, without any overall gain change.
 
Hmm, I would no limiting what so ever.

You should asy your ME what he or she would like.

If you have not choosen a ME yet. I would mix it so that no peaks clip. Any compitent ME will know how do deal withanything you give him as long as it's not excessively quiet, or clipped.
 
You should asy your ME what he or she would like.
Click to expand...

true...true

I put a very very light limiter, and by light I mean light...just enough to mildly tame any peaks that might jump over -3....that's just how I do it, and I am not endorsing this technique...

Every ME I have ever sent anything to requested 3 db of headroom, so as I rule that's how I've always mixed, and have always had albums mixed.

the same can be said for someone mixing with the master bus hitting -6db...the ME can just as easily bring up volume, as he can bring it down.

The professional rule of thumb is -3db IMHO... :cool:
 
blueroommusic said:
true...true

I put a very very light limiter, and by light I mean light...just enough to mildly tame any peaks that might jump over -3....that's just how I do it, and I am not endorsing this technique...

Every ME I have ever sent anything to requested 3 db of headroom, so as I rule that's how I've always mixed, and have always had albums mixed.

the same can be said for someone mixing with the master bus hitting -6db...the ME can just as easily bring up volume, as he can bring it down.

The professional rule of thumb is -3db IMHO... :cool:
Click to expand...
Yes, -3db is a great target.
But I think it's best to achieve that volume through the actual mixing process. Even if it means bring the master fader down, or up.

As you said a ME can always bring volume up or down. But there is little he can do for a limited transient. that is the problem with the limiter.
I understand the mindset of a limiter. It's better to be limited than to be clipped. that is without a doubt. But if you have the better option of smply turning it down, I'd do that. Most DAW software or amplitude plugins will let you "normalize" to a specific level. If you punch in -3dB in a normalizer, your highest peak would just tap -3db. And there would be no deformation of any waveforms (other than the reduction of amplitude of course).
 
What is "headroom", is it where a mix peaks at or the difference between the peak and RMS values (aka crest factor)?

As long as there is no clipping in the mix you are fine. Taking the master fader and reducing your mix 3 dbFS isn't anything magical that a good ME can't do. If you're sending 24 bit files, there's very little in the way of degradation of sound. If you're sending 16 bit files, it's actually better to get closer to full scale (approx 1-3 dbFS below peak).

The real issue with mix "headroom" is when the crest factor is small and an ME doesn't have room to work with compression/limiting. I'll take a mix with a good crest factor that peaks at -1 rather than a low one that peaks at -6 or less any day.
 
Last edited:
I'm big on unobstructed headroom - NO limiting. NO mormalization. And always err on the side of "too quiet" - Which is nearly impossible anyway.

I had a mix come in a few months ago that for some reason, was down 48dB. The PEAK was below -48dBFS.

The engineer was here, he didn't know what was going on and said that he'd go back and redo it.

Either way, I turned it up around 45dB and it sounded just fine. Of course, I had him redo it anyway (if for no other reason than to find out what went wrong at the output) but it would've been *totally* usable even if he couldn't redo it.

As a "rule of thumb" - the best sounding mixes that come in here are riding "around" to "well below" -20dBRMS. That's where the "professional" engineers (at least the vast majority of the ones I deal with) are are mixing - which makes perfect sense if you figure that to be around 0dBVU. In the UK, I belive it's become "the standard" -

Once you worry about your mixes "being loud" is where you usually do something to all but insure that they WON'T be when they're done.

Just mix the freakin' thing and *don't worry about the loudness* - Whether I'm mastering it, or Tom, or yourself, or anybody else. Just do what the mix is telling you to do and make sure you're not clipping. Personally, I try to leave *at least* 6dB of headroom - But if I'm getting the bulk of the mix to sit around -24dBFS or so, that usually takes care of itself.
 
Thanks for the replies, i think the real question that i wanted to ask was...

Once i have lets say 3 Db of headroom ....With the plug ins i have available to me ....i.e. Waves L2, Izotope Ozone, and PSP Vintage Warmer ...

How should i apply these plug ins to the stereo mixdown in order to get the best possible result...

I am producing a 3 song Demo Cd for a friends band... They are not looking the cd mastered professionally as this is outside of their budget ...

Any advice or suggestions are welcome.

Thanks :)
 
Without knowing much about those plugs, I can only say generally that is very dependent on the music, but limiters will tend to be last, otherwise a subsequent plug could push you into clipping if you aren't careful.
 
Please see:
Headroom

If you have 3 db of headroom your mix is already too loud.

To answer your question though, what do you think the mix needs?

From the plugs that you mentioned I would guess that the L2 is a good candidate. Use this to get the mixes to the level that you want for what is appropriate for the songs (basically reduce "headroom").

In regards to the other plugs, only use them if needed. Without hearing the music there's no way to give a good recommendation.
 
I may be moving off topic a bit here, however...

I really love albums produced in the 60's and 70's....I love the dynamic range that is really REALLY noticable when you listen to songs from albums back then, compared to albums from the last 10 years.

One Song that I really Really love listening purely for the fact that you can actually feel the change in dynamics throughout the song is Wild Horses ( acoustic Version ) by Rolling Stones.....

I stuck the latest U2 cd in straight after listening to that song and checked out the Stereo Wave file for Vertigo....

I must have sat with my mouth open for about an hour ...

:eek:
 
Now wait a second, let's see if I get you right here.

Here's a scenario. I am recording a rock band. I try to get a decent level to disk so that nothing clips, but the signals are nice and strong. I then go on to mix this recording in the box trying not to stray too far away from unity on each of the faders, since the resolution of the fader is best in that area (and maybe to avoid too much of a rounding error, if I'm being even more geeky). In this case, I'm sure to get overs on the mix bus, at least each time the transients of the kick and maybe the bass guitar really fall on top of each other. So I pull the master fader down just enough, so that the peaks happen within the last dB before clipping, e.g. -0.7 or -0.3 or whatever.

So, what you're basically telling me is, I should bring down the level by a good amount, lose resolution on the way (yeah I know, it's 24 bits and I shouldn't worry about it, cause even -48dBFS in 24 bit equals 16 bit resolution), just so that the ME can bring it back up again?

Wouldn't it be more sensible to give the highest possible resolution to the ME?
 
nessbass said:
So, what you're basically telling me is, I should bring down the level by a good amount, lose resolution on the way (yeah I know, it's 24 bits and I shouldn't worry about it, cause even -48dBFS in 24 bit equals 16 bit resolution), just so that the ME can bring it back up again?

Wouldn't it be more sensible to give the highest possible resolution to the ME?
Click to expand...

The general idea that John and I are trying to get across is that you want to preserve as much of the dynamic range as possible in a recording. Once lost, it can't really be restored to it's original form.

In regards to peaks, the concept is to try to reduce any processing that can be avoided. For example, when an ME gets a mix, often the first stage of processing is EQ. If you are boosting one or more frequencies, there will be overs if the original peak level is high and as a result the level will need to be lowered before EQ to prevent it. You can avoid that gain processing if the original mix is a few db beneath where the final master needs to be. When mixing, the master fader (at least in Pro Tools) is doing the reduction in 48 bit (in TDM systems) before outputing a 24 bit file, so the reduction would be better done there than later as a 24 bit file.

As I said in my earlier post, the peak reduction is trivial compared to the damage that some folks do in regards to dynamic range. That's where the real emphasis should be.
 
nessbass said:
Now wait a second, let's see if I get you right here.

Here's a scenario. I am recording a rock band. I try to get a decent level to disk so that nothing clips, but the signals are nice and strong. I then go on to mix this recording in the box trying not to stray too far away from unity on each of the faders, since the resolution of the fader is best in that area (and maybe to avoid too much of a rounding error, if I'm being even more geeky). In this case, I'm sure to get overs on the mix bus, at least each time the transients of the kick and maybe the bass guitar really fall on top of each other. So I pull the master fader down just enough, so that the peaks happen within the last dB before clipping, e.g. -0.7 or -0.3 or whatever.

So, what you're basically telling me is, I should bring down the level by a good amount, lose resolution on the way (yeah I know, it's 24 bits and I shouldn't worry about it, cause even -48dBFS in 24 bit equals 16 bit resolution), just so that the ME can bring it back up again?

Wouldn't it be more sensible to give the highest possible resolution to the ME?
Click to expand...

I imagine pulling down the master fader the way you did is going to sound just fine. I like to keep the master fader at 0db but I don't think it matters much.
 
lurgan liar said:
I may be moving off topic a bit here, however...

I really love albums produced in the 60's and 70's....I love the dynamic range that is really REALLY noticable when you listen to songs from albums back then, compared to albums from the last 10 years.

One Song that I really Really love listening purely for the fact that you can actually feel the change in dynamics throughout the song is Wild Horses ( acoustic Version ) by Rolling Stones.....

I stuck the latest U2 cd in straight after listening to that song and checked out the Stereo Wave file for Vertigo....

I must have sat with my mouth open for about an hour ...

:eek:
Click to expand...

Yes, 'tis true. But whenever somebody has told me lately that they wanted a dynamic, classic, analog sound, it later turned out what they really wanted was crushed to death :(
 
masteringhouse said:
When mixing, the master fader (at least in Pro Tools) is doing the reduction in 48 bit (in TDM systems) before outputing a 24 bit file, so the reduction would be better done there than later as a 24 bit file.
Click to expand...

Okay, that makes sense. Thanks.
 
nessbass said:
Here's a scenario. I am recording a rock band. I try to get a decent level to disk so that nothing clips, but the signals are nice and strong.
Click to expand...
There's another *possible* problem there -

Once everything is ITB, if it's going to indeed stay there, a lot of things go out the window to a point.

But it's BEFORE the box that's going to make a drastic difference in the sound quality - Whenever I see "strong signal that doesn't clip" I think "as loud as possible without clipping" which is really, really bad form, and depending on the front end, can be quite damaging to the audio.

I've come up with a new rule of thumb - "You're probably tracking too loud."

Okay, maybe not YOU, but wow... Preamps are made to sound best at or under 0dBVU. A lot of people want to "record hot" and literally *overdrive* the preamp (not unlike overdriving a Marshall stack's preamp) to get the signal up into those last few precious bits.

Those last few precious bits are there to get BETTER sound at NORMAL levels. -0dBFS is NOT a "normal" level - Whatever happened to headroom?

I get a lot of projects in from a lot of different types - Hobbyists, professionals and anything in between. Whenever I ask a question or two, it almost always comes up somewhere -

The professionals, with the great sounding mixes, track using their gear the way it was designed - "Oh, somewhere around -20dBFS or so" is the usual tracking level. And around -20dBRMS is around the usual mixing level. This is typical with the mixes that come in here sounding wonderful. The open, airy, clear, focused sounding mixes that have huge headroom and intact dynamics.

No surprise, these are also the mixes that wind up coming out MUCH louder in the end - They can handle the "abuse" much better than:

The "pinched" and "unfocused" sounding home-brew stuff that's tracked "as hot as possible without clipping" - which is anywhere from maybe 16 to 22dB HOTTER than the preamp is designed to run. That's a *serious* amount of signal to be shoving out of a preamp. There's a lack of focus, added distortion, messed up imaging, a whacked out signal to noise -

The really great sounding home-brewed stuff just so happens to be tracked and mixed the same way as the professionals... They might not even know why. "I was just trying to play it safe" and "I saw something on a forum about not recording hot" are popular replies.

And it really puzzles me still on this... This is "day one of class" stuff. But it seems like a huge percentage of people skipped the first day of class for some reason...
 
So that's maybe the reason why is "mismatched" my mixer and soundcard in the past. I used the pres in a mixer that is designed to output levels at +4dB, but I ran the soundcard's converters at -10dB, so that the levels showed up in the meters in Logic. These meters only start to display anything if the signal is above -20dBFS, which always gives me the feeling of tracking too low, if they'retracked "correctly". Is there any problem in doing this that I'm not aware of?
 
You must log in or register to reply here.
Back
Top