MIXING IN CUBASE 0DB IS A RULE FOR THE MASTER TRACK??

[fernando]

Hi !!!
I´m currently reading Tweak´s guide about Mixing in the Virtual Realm of the Sequencer...I produce pop music, with bits of house and electro and I use Cubase SX 3 and Reason.

I have a question about Mixing that´s been on my mind for ages...I know individual audio tracks shoud never exceed 0db but my master track always goes over 0, since I´m putiing all the tracks together...

I know producers that leave the master at +7db and use a compressor or limiter, and I know people that say that the master also should never exceed 0db and in order to make your sound louder you can use plug ins like L1 Ultramaximizer...

Who is right? The first guys say it´s better to excced 0db and compress afterwards (without hearing any distortion of course) because they don´t want to lose WEIGHT , they want FAT sounds...they´re mainly electronic music and hip hop producers and they say that´s what people like the neptunes do, how modern pop music producers get that loud sound...

Who shold I trust? Is there a rule for that? Or it´s all about modus operandi, each person has one? Or I can just follow my ears, if I don´t hear distortion then fine?

Thanks a lot guys , I hope someone can give new ideas and insights on this
Fernando

[epic]

stay away from that 0db mark. far far away. If it needs to be louder, turn up the amps. pumping and compressing the whole song makes it dull and lifeless. Loud, sure, but lacking of dynamics and range. Don't kill your song using a compressor or limiter when someone else is going to do just the same to it later and kill it more.

[MASSIVE Mastering]

I know individual audio tracks shoud never exceed 0db but my master track always goes over 0, since I´m putiing all the tracks together...

It's VERY likely that you're recording your tracks much too hot.

I know we go over this a lot, but it tends to slip past an awful lot of people...

When recording your individual tracks, use 0dBVU as a reference level. Somewhere, somehow, someone goofed up and started using 0dBFS as a reference - It seems that started what must be a decade of confused people running around not knowing the levels that all of their gear runs at (which is, and always has been, 0dBVU).

There are SO many positives to this, not the least of which are (A) MUCH better sound quality at the input and (B) mix levels that almost take care of themselves. There are NO negatives - Especially in 24-bit. There's just no downside.

Wait - Maybe there's... No... Sorry. No downside. None.

I'd actually like to find the guy who first said "Just turn it up until it almost clips" and give him a good smack...

[chesterdesmond]

I have a question about Mixing that´s been on my mind for ages...I know individual audio tracks shoud never exceed 0db but my master track always goes over 0, since I´m putting all the tracks together...

I've also wondered about this myself...I have a Delta 1010LT (8 in/8out) and I am routing all the tracks separately to my mixer where I am setting the levels; none of the levels are over 0db, in the software (Tracktion) or on the mixer. However the master level in Tracktion is peaking way over 0 (if master volume indicator is in 'peak RMS' mode but not in 'average' -maybe vice versa). I am routing the audio from the mixer back to the ins and recording a new stereo track of my mix. Using this method, can I safely ignore the master levels in Tracktion?...to my way of thinking, none of the audio is being routed through the master bus of Tracktion (except internally) even though (to add to my confusion) I can alter the volume level with the main (software) fader, which I assume is a global volume for the Delta. Would these master levels only come in to play if I was bouncing the track down in the software?

MM posted while I was composing, so if anywhere I seem to be ignoring his remarks, that;s why....but it does lead to me asking (not being at home)...are the 0 levels in the software at a different 0 db than on the mixer?? I know I've read this before but I like the way MM sighs

[owel]

>are the 0 levels in the software at a different 0 db than on the mixer

You must calibrate everything, i.e. the signal chain before you start worrying about 0dB. Otherwise, what you're seeing may be meaningless.

0dB FS at +4dbu will be 1.22 Volts RMS approx. (make sure to use an AC rms meter).

0dB FS at -10dbV will be 0.31 Volts RMS approx.

[chesterdesmond]

ok ... any ideas on a link or a tutorial of how to do this..ie what all needs to be checked...
by calibration do you mean to follow the signal all the way through the path and each permutation of that path and ensure that +4 is set on each piece of hardware?

[fernando]

thanks for the answers!!

especially for producers that are going 100% digital and need no calibration, do those rules apply?

i still have doubts on this, i´d also love to see a lesson on this...
i wonder how britney spears get that loud and fat sound on toxic (just an example) working under 0db??

cheers!

[fernando]

do you guys know where I can read about dbfz or dbvu???

[mcatalao]

do you guys know where I can read about dbfz or dbvu???

You have software converters (dbu to dbv to dbfs...) on the net... Just google it!!!

Anyway, i find myself trying not to overload my master track in Cubase, so that everything sounds clear and don't have saturation issues.

I play light pop and traditional music, so my tunes tend to be very dinamic.
One usual problem is that people tend to go near 0db on every channel, and also, in each send or insert you use, the level of that particular channel will be modified.

I believe Bruce or John can be more technicall, but my perception of this is that all channels will be summed (of course at the same frequences i guess) generating an overloaded master track, wheter it is on some specific frequences or at the overall mix. But the meter doesn't say to you wich are the frequencies saturating your master track, so you have to understand wich tracks, or group of tracks are doing that...

Most of the times the master will peak due to the most dynamic instruments like drums, so i try to add compression in this instruments and not using at the master track...

But when i can't do nothing about this... I use a limiter the master at 0 db... But the result isn't quite good as being carefull at the mix stage...

Finally i look at compressing the master track like a process that will bring the level of the master UP, and at the same time shortening it's dinamic range. But wouldn't that be a job done in the mastering process, along with the other things (multiband compression, eq, subbass threatment, etc, etc...)???

Maybe i'm saying something STUPID??? So please correct me if i'm wrong...

[chesterdesmond]

Finally i look at compressing the master track like a process that will bring the level of the master UP, and at the same time shortening it's dinamic range. But wouldn't that be a job done in the mastering process, along with the other things (multiband compression, eq, subbass threatment, etc, etc...)???

I think that it is considered smart to leave some head room for the mastering engineer to work in, ie if you send in your master at 0 (s)he can't do much with it, especially volume-wise. So I guess to save me asking this in a couple weeks, does someone know a general rule of thumb as to what a good final level is for something that is to be mastered?

[fooman]

I dunno where I read this, but -12 is a good level to shoot for. I could be very wrong since I obviously read it in an internet article, but it works for me since it allows me to have waaaay more room to compress/limit my projects.

[mcatalao]

Well, if you read me correctly, i don't agree with compressing the master track (and was questioning, why compresse the master, once you would do that in the mastering process?).
Anyway, i agree with you.
I think you have to mantain some dinamics (geez, even the final product HAS to have some dinamics!).

Though, LIMITING, will alow to controll peaks, so you don't overload your file when you export it from cubase.

What do you mean by " good final level"? (average mix level? Max Level? Peak leve? RMS Peacl Level??? There are so many things you have to consider... )

IMO, the final mix level should be one that allows to hear all instruments in their dinamic differences, and still have the max headroom possible. Of course that will depend on the type of music.

I.E. if you're playing classicall music your mix should have a lot headroom... Really A LOT.

After mastered, most of classical CD's won't be LOUD! Well, they might have hard peaks. But they still have those 60 to 70 (or more) db headroom.... And at that point (Please correct me if i'm wrong again, John (massive mastering)), i don't believe the master engineer, had a lot of work where loudness matters.

In other hand, if you're mixing a pop song, not only the instruments have less dynamic range as most of the time, you will compress much of the elements (usually the more transitive ones, like rithm instruments, voice, etc...).

So your mix will be less dinamic. But less dinamic won't mean louder, or softer...

[chesterdesmond]

Well, if you read me correctly, i don't agree with compressing the master track (and was questioning, why compresse the master, once you would do that in the mastering process?).
Anyway, i agree with you.

and I was agreeing with you just reinforcing what you were saying really.

What do you mean by " good final level"?
IMO, the final mix level should be one that allows to hear all instruments in their dinamic differences, and still have the max headroom possible. Of course that will depend on the type of music.

My understanding is that headroom is what is available above the highest peak..I am mostly talking final volume level, so I guess my question would be: how much headroom should we leave above the highest peak for the mastering engineer to work with?

[mcatalao]

Well my perception of headroom is a little different than yours...
I kind look at it as the difference between the highest peak in your mix and the lowest. Basicly it will be your total available dynamic range.

From wikipedia:
"Headroom is a term used in signal processing (audio, data, etc) to refer to the maximum allowable signal level before compression or distortion.(...)"

http://en.wikipedia.org/wiki/Headroom

They also have a great example where you can see headroom differences between severall audio systems.

Anyway, your question... I don't think you should think it like that, because you might have peaks near 0 db in you mix... and at the same time it can be very dinamic!

[MASSIVE Mastering]

The digital headroom thing allows for processing - it's the ANALOG headroom that makes the HUGE difference in recording quality. That ends up translating to digital headroom (which is exactly how the system was designed to work - No accident there).

Take a piece of gear - Anything - Preamps, EQ's, compressors, limiters, pee-wee-harmanizers, CONVERTERS, anything. That piece of gear is designed to run at 0dBVU. That level equates to *roughly* -18dBFS plus or minus a dB or two.

Headroom in the gear is the the difference between that level (0dBVU) and the level where it will clip. With most gear, that's going to be somewhere between +20-25dBu - Several dB *above* digital full-scale (-0dBFS).

But just because it won't *clip* by the time it hits full-scale doesn't mean the sound is going to change dramatically on the way... When the output knob is at unity and the input signal is raised to 0dBVU, there is a certain amount of noise that will be nominal and normal to the unit (and as we've just gone over, that level will read somewhere in the neighborhood of -18dBFS on the digital meters).

When you raise that level to get "hot" digital levels, you're *amplifying* it and taking it out of the nominal range where it wants to work. You're not just turning up the signal - You're not just turning up the noise - You're making the noise *worse* as the level increases in many cases. The headroom that's there to support transients is now being used to support the entire weight of the signal.

Wow! I just looked at that Wikpedia article (and bookmarked it) and there's a VERY telling image with the phrase "Master recording assumes 24-bit with 24dB of headroom assigned."

*24dB* of DIGITAL headroom is the Eurpoean standard. And look at the "Compact Disc" bar - Boy, if that doesn't look like 18dB, I don't know what else does...



Couldn't have said it better myself... Although I rarely ask for more than 18dB...

Anyway - Back to the grind...

When you record a bunch of tracks at these *normal* levels (this is how it's been done "downtown" since day one) and you toss the faders up to start a mix, YOU are in control of it. You might not clip anywhere in the mix depending on the density and number of tracks even with all the faders at unity. Where if you recorded tracks at hotter levels, you'd have to turn them all down and start a battle of levels just rying to figure out where to start everything.

And absolutely - As much as I hate to say it, if you're looking for "sheer volume" in the end result, THIS is how you get it - Headroom - On every track, on ever piece of gear in the line, on every group, on every main buss... Mixes that have been "treated well" along the way are FAR more likely to handle the "abuse" at the mastering stage. Plus, it's very likely going to sound worlds better in the first place. Plus, this is how it's been done for decades. There's a good reason for that.

[mcatalao]

John, 1 question...

I think i'm confusing headroom and dinamic range... It is not the same?
(i guess my last post is a little stupid hã?)

If not, what's the difference and how are they related?

Thank's in advance...

[owel]

John, 1 question...

I think i'm confusing headroom and dinamic range... It is not the same?
(i guess my last post is a little stupid hã?)

If not, what's the difference and how are they related?

Thank's in advance...

No, headroom and dynamic range are not the same. But there's correlation between them.

Dynamic range is Headroom + everything between 0dB and your noise floor.

Headroom is from 0dB up to the max your equipment can handle without clipping... good gear have it at +24dBu. So-so gear at +18dBu.

[JohnnyDemonic]

ok, so I have cubaseLE, a firepod & a behringer mixer... signal goes into the mixer & I PFL that channel & get the signal to hit 0db (or just below 0) at the peaks... I adjust the Firepod gain so that the peaks in Cubase peak at 0db (or just below 0)... this is maximizing the headroom on the board & also maximizing headroom in Cubase... doing this should keep the master channel below 0db & give the mastering house plenty of headroom to work with, correct?

[Dj DynaMix]

I know producers that leave the master at +7db and use a compressor or limiter, and I know people that say that the master also should never exceed 0db and in order to make your sound louder you can use plug ins like L1 Ultramaximizer...

Who is right? The first guys say it´s better to excced 0db and compress afterwards (without hearing any distortion of course) because they don´t want to lose WEIGHT , they want FAT sounds...they´re mainly electronic music and hip hop producers and they say that´s what people like the neptunes do, how modern pop music producers get that loud sound...

This part REALLY interests me...I use to think staying away from 0db (and then getting back up there with a limiter) was the right move - but I too have now met producers who do no type of limiting and almost ALWAYS mix the track down at some volume over 0db --- and it sounds awesome... much smoother and "more together" than my tracks...has me wondering if I am doing too much work for no benefit...

[Farview]

ok, so I have cubaseLE, a firepod & a behringer mixer... signal goes into the mixer & I PFL that channel & get the signal to hit 0db (or just below 0) at the peaks... I adjust the Firepod gain so that the peaks in Cubase peak at 0db (or just below 0)... this is maximizing the headroom on the board & also maximizing headroom in Cubase... doing this should keep the master channel below 0db & give the mastering house plenty of headroom to work with, correct?

Not quite.
0dbvu on the mixer should be -18dbfs in Cubase. You don't want to use up all the headroom.

0dbVU on the mixer is line level. -18dbFS in Cubase is line level.

Software and hardware meters are using different scales and they are different types of meters.

VU meters are slow and measure average level.
Digital meters are fast and measure peak level.

That is where everyone gets confused. The two different types of meters measure different things in a different way. 0dbfs is not 0dbvu

[MASSIVE Mastering]

ok, so I have cubaseLE, a firepod & a behringer mixer... signal goes into the mixer & I PFL that channel & get the signal to hit 0db (or just below 0) at the peaks... I adjust the Firepod gain so that the peaks in Cubase peak at 0db (or just below 0)... this is maximizing the headroom on the board & also maximizing headroom in Cubase... doing this should keep the master channel below 0db & give the mastering house plenty of headroom to work with, correct?

To further that - If you're talking about the *VU* meters (ladder bars? Anything?) on the BOARD, then that's probably dBVU. That's fine. Hitting zero occasionally is maximizing the headroom at the board.

That it - Done. Everything else at unity, you should be hitting around -16 or -18dBFS in Cubase. That's a good thing. Turning the gain past unity is *using up* all that headroom.

Quoiting myself -

When you raise that level to get "hot" digital levels, you're *amplifying* it and taking it out of the nominal range where it wants to work. You're not just turning up the signal - You're not just turning up the noise - You're making the noise *worse* as the level increases in many cases. The headroom that's there to support transients is now being used to support the entire weight of the signal.

[NoShame]

man... I am so confused.

John are you saying that if my peaks are maxing out at 0dbVU on my board, then routed into my software (Cubase SL3) that those peaks (at 0db on my board) translate to -18dbFS in Cubase??

If I do the above - and I look at my Cubase display... it also looks like it's peaking out at 0db. What the heck am I missing here???

Gonna start reading this thread from the top again... <sigh>

[mcatalao]

Ho what a mess (my brain)...

And this values you gave are different for -10 dbv devices and +4 right???

[Farview]

If I do the above - and I look at my Cubase display... it also looks like it's peaking out at 0db. What the heck am I missing here???

0dbvu on the board is where the green leds turn to yellow ones, not the red ones. Look at the scale.

If you are reading the meter on the mixer correctly, then you are adding gain at your interface, or you are sending a +4 signal to a -10 input.

You have a gain staging problem.

[MASSIVE Mastering]

mc - the values should be the same as long as the gear is hooked up properly (no +4 signals going to -10 inputs, etc.).

But No Shame - There's got to be something goofed up somewhere...

I'd start from the top down (the preamp on the board) and make sure that *everything* on the way to the DAW (including the DAW - I recall that Cubase has input gain trims, yes?) is at unity.

Set up the mic and do what you do - Adjust the pre-gain so that the levels on the board's meters are hitting 0dBVU. Or even better, run a CD player with a stereo pink noise track into it so you can concentrate on everything else with a steady signal. Everything else at unity - The main output, the channel gain, everything at the DAW. You should read somewhere between -20 and -16dBFS.

And yeah, make sure you aren't stuffing a +4 signal down a -10 pipe... That would add an awful lot of gain (about 12dB) and if everything else is actually set properly, it might not even add all that much distortion. So, it *could* possible go unnoticed...