I have a very simple question about the levels in mixing, here it is;
I know the peak of the final mix-down must be somewhere between -6 and -3 db, let's say it will be 6 for this example. I start by recording every instrument around 15 and 12, adjust it on my MOTU audio interface's input level knob according to the [="http://www.apple.com/logicstudio/"]Logic[/]="http://www.apple.com/logicstudio/"]Logic[/]'s level meter. So now I have 5-10 tracks with different instruments. Later on, I think the most loud part of the song should hit -6 while the more quieter parts -10 or around and here arises the problem. How can I adjust a track to hit max -6 db while controlling it's dynamic values? should I just put a limiter ([[url=http://="http://www.waves.com"]Waves[/]="http://www.waves.com"]Waves[/] L1 perhaps) and hit the compression until the sound starts to get damaged. But on the other hand I don't think it's normal to have a limiter on each track on a mix. I wonder if there is a normal legible way of controlling the levels of tracks?
Many thanks and warmest wishes,
You're concept is incorrect. You mix so that every thing sounds the best it can. After you are done mixing, then you worry about bringing the entire level up. But, while you are mixing don't fuss yerself with it. Also, the best way to adjust levels is to pull sliders down and not to push them up.
Agreed. With 24 bit resolution in the digital realm this almost doesn't matter. In the analog world it is a different beast.
Hello, thanks for the replies. Yes I know the concept of mixing, I've done that. But now I have to prepare the mixes for the stage of mastering and I want to have control on the dynamics of each section on each track as I said. So when I have the bounce I want to see the loud parts hitting -6 and quieter parts around -10 as I said. I still how can I achieve this without actually putting a limiter on each track? Or what is the proper way of finalizing a mix with levels.
If your output levels never peak over -6 to -3 db you'll be fine. I would totally forget about the db peaks of the quieter parts.
Thanks. Yes, that was just an example, what still worries me is to create the loud parts, controlling the headroom and loudness in a proper way. Should I put a limiter and hit the gain on each track so none will exceed -6 and still have the amount of headroom that I want and have a nice overall sound?
Really by doing this you are defeating the purpose of mastering. I would say generally that putting a limiter on each track for this purpose is extremely bad practise.
In order to preserve the sense of the songs dynamic the mastering engineer will employ a plethora of tactics to let the song breathe. Your job is to give the mastering engineer your vision of the song with what ever dynamics you see fit. The mastering engineer is not there to squash the hell out of it. They are there to balance your songs to fit together, sound great on all systems and be reasonably loud for the genre.
Ailgun, listen to your song at very low level. If it sound good and you ear every instruments, don't touch anything and send it to the mastering.
If at low level you are loosing an instrument for the whole song : work the eq and/or the main level of this instrument.
If your loosing an instrument just at some places in the song : Try to figure out if it's being mask by another instrument's dynamic or it's own dynamic. And cure it with a lite compressor (2:1 at first).
Mixing is all about fixing problems and being carefull about creating new ones ;)
Hello, thanks to all for the answers.
First of all I want to say I understand your statements but please let me explain myself. Firstly I'll be doing the mastering for this album (for vinyl actually) and I believe such great mixes doesn't require much in mastering. So having these vast songs with many different sections somehow push me to have a very solid mix before I go to the mastering stage.
Here I've uploaded a track of ours to Soundcloud, you can see the overall waveform. No normalization of course, everything just as recorded and a very very rough mix is done. It's very quite.
But the problem is that even the middle part -where I want it to be considerably loud- has a vast headroom and isn't loud. I can give up on that but I don't how? Technically how can I reduce the dynamic range on the middle section of the track, make it louder -making it peak at -6- and make it seem like this;
[MEDIA=soundcloud]nineinchnails/came-back-haunted-2013 (This doesn't seem to have a sparse headroom but still sounds really good) Should I find each tracks peak, let's say -12, and increase it by -6, and if it isn't loud enough compress it and find the new peak and increase the gain again? Or what? I don't think this is the way I should hand this song to a mastering engineer or start mastering it by myself.
Also I adjust my levels according to the Logic's MultiMeter, I wonder if it's the right method or should I also learn to use VU meters or Loudness Meters?
Mastering for vinyl is a specialty in this day and age. It requires very specific eq profiles (RIAA) and levels. This is not something for a beginner to try. Also, you will still need actual mastering. To think otherwise because it is vinyl is flawed thinking.
Now as for the finalizing or even the follow on mastering, you do NOT limit individual tracks in a misguided attempt to bring the overall level up. You mix the tracks-which can include individual compression or expansion-so the overall mix sounds good at both very low levels and loud levels. Once the entire mix sounds good then the finalizing and mastering can occur. One element of this is indeed making the overall volume able to compete reasonably with other productions in the same medium. That overall volume is decidedly different for vinyl versus digital compact disc or DVD or 1 bit DSD.
Basically, you are eager but need to get more educated in the process before you are ready to "master" or more accurately, "finalize" your mix. Mastering is more than you think.
Hello, thank you for your answer. The thing is that I actually am aware of all the things you said theoretically but fall short on the application. Lets say I'm pretty satisfied with the mix on this link;
This has a waveform like this on the Logic and below I have an ideal mix with overall loudness and everything.
So what I'm trying to achieve is that to realize what the final db for an mix should be before it moves to mastering? As far as I read it should be around -6 so it gives some space to the mastering engineer. So how can I adjust the finalized mix of my song peaking at -6 with a similar perceived volume compared to other examples in same medium?
Lastly, I realize mastering for vinyl is a big issue, but in my country (Turkey) I'm sure there is no one who can do it. Would you suggest getting it done online?
ailgun, post: 406875 wrote: Thanks. Yes, that was just an example, what still worries me is to create the loud parts, controlling the headroom and loudness in a proper way. Should I put a limiter and hit the gain on each track so none will exceed -6 and still have the amount of headroom that I want and have a nice overall sound?
Get this obsession with limiters in the mix out of your head. I use limiters all the time in mastering but I can't remember the last time I used one in a mix. Just make the mix sound good while leaving some headroom.
If a section is too low then use editing and clip gain or volume envelopes to make it right.
First of all, you don't quite have all the theory understood. The difference between something that has not yet been crushed and something that is mixed nicely is not the PEAK dBV. What makes things on the radio sound "loud" and just plain ugly and nasty is the LACK of dynamics created by raising the RMS averages as high as one can make them. Most pop musicians take this step way too far.
Raising the RMS value has nothing to do with limiting any individual track whatsoever. Make the mix as pleasing as possible and bounce it down to a final stereo .wav or .aiff file. Once you have that final mix completed in stereo waveform is when you begin trying to "make it louder" which is a whole other bag of cats.
You absolutely DO NOT try to "master" or raise the "loudness" while working on a multitrack mix.
So, looking at the screenshot I put above, can you tell the mix (asuming it sounds right to me) is ready and can go to the stage of making it louder? And I know how incresing the volume makes it less dynamic by reducing the headroom.
The wave form on top is plenty "loud enough" to send to mastering. To finalize for "volume" you are raising the minimum levels while compressing the top peaks down. The compression is what creates the monotone volume that you are forced to hear on the radio. When it looks like a brick like the bottom wave form then it sounds nasty though in pop music it is somewhat required. The trick is to make it the least amount of nasty to loud ratio that you can.
Thanks, I see. Yes I know the purpose and result of compression.
Can I ask a few more questions, if you would mind to answer.
How does my mix sound overall? It's very rough but I can take any advices upon that.
Would you suggest getting the mastering done by someone online, or would you advice to try to learn it in a few months?
One last thing, on my mix on the middle part you can see a percussive sound way over the rest. I wonder if on the mastering whence the rest on the below is raised that percussive sound will be swallowed?
I had a neighbor, she was good for swallowing my percussive sounds !! thumb
Mix your songs as you want to ear them, don't do any ajustment thinking the mastering might hide it or turn it wrong. A good mastering engineer, will enhance your song not destroy the work you put on it.
Thing is, you might say that because you tried to master without the proper tools or knowledge. Please don't say I know again.. I read the entier thread. If someone knows how to drive a car he won't ask if he can drive wrong side of the road and then say I know what wrong side meens..
One thing that might have be wrong doing is to put an instrument very clear and let say the vocal very dark.. at mastering, we might want that vocal to sound clearer and so the other instrument will be too clear.. A good mix won't have that kind of problems..
Hello pcrecord, thank you very much for your reply but the thing is that I know everything you speak of, even the neighbor you had, haha just kidding :tongue:
Thank you it was really enlightening, now I see and feel much more comfortable. But a new question wanders around my mind now is the discussion of headroom of mix... Let's say my mix peaks at -10 dbFS at those loud percussive sounds peak, can we say the headroom of my mix is 10 dbFS since the mastering engineer will have that much of space to work on?
Also within the multi-track file I used to think that the darker blue area between the RMS and peak would represent the headroom, is that correct after all?
Here you can see the meter.
In the digital world there's only one LAW, you can't capture a signal above 0db because you'll get crapy noise artefacts. That 0db is the one you see in you DAW when monitoring the input of a track.
The common practice, is to record at way lower levels to be sure you'll never cross 0db. I record signal aroud -12 with ocasional higher peaks that can go to -9db or -6db maximum. Since I'm not using compression while tracking, If a sreamer ever surprise me and hit 0db, I stop the recording, ajust the preamp and ask the customer to start over that part.
Looking at the wave form image in your daw is a good gizmo to help making copy and paste and other track manipulation (ex : it's easier to make volume automatisation when you see where the notes are...) In sonar you can even accentuate those wave images without affecting the level of the track (if the track was recorded at a low level)
At mix time : The only important thing to remember when mixing is to check the master level. It is pointless to check the wave form images to try to figure headroom and other signal level estimation. Of course any track should also not peak if effects are added to it. The master level of a final mix should go around -12db to -6db. Why not going for -1db ? Simple, A lot of transients are too fast for registering on your DAW VU therefor, you might exceed 0db without knowing it. Plus, some DAW have protections from exceeding 0db and use a hidden drastic limiter to correct the output AND some DAW have analog emulation that will create distortion if the main level is too hot.
ALSO, remember, you can send a -24db to mastering and it will be fine. It's the mastering job to get your song to an industry standard level matching the musical style.
Your job (and mine) is to avoid creating problems. I spoke of the EQ frequencies in my last post. I should also speak of volume levels. If 2 instruments hit at the exact same point in time and you mix one too loud (as the percussions you spoke about) it is very understandable to say that if I apply a limiter or a fast compressor. Both will go down and than we might loose one.. A well balance mix won't have that kind of problems..
ailgun, post: 406909 wrote: Let's say my mix peaks at -10 dbFS at those loud percussive sounds peak, can we say the headroom of my mix is 10 dbFS since the mastering engineer will have that much of space to work on?
You leave plenty of headroom so you can make adjustments without the slightest risk of clipping. The mastering engineer can work on anything that isn't clipped, but maintaining headroom is just good practice throughout the process. Drive down the center of the lane, not right up against the guardrail.
ailgun, post: 406909 wrote: Also within the multi-track file I used to think that the darker blue area between the RMS and peak would represent the headroom, is that correct after all?
Here you can see the meter.
No, that's just the difference between the peak and RMS at any one moment. Headroom is the difference between the highest peak of the song and 0dBFS.