Preparing your tracks for professional mastering...

[safeandsound]

Hi all, I thought I would post up a basic guide to prepping audio files for mastering which many people appear to have trouble with, it is very straight forward.

If you are planning to send your music for mastering or even if you are "self finalizing" your own tracks DIY style, it is a good idea to plan the finalizing session seperately from your mixing. Remove your output bus limiter, when you remove a limiter do consider that this might mean that the signal level could easily clip the master bus output. (this means the signal level has exceeded digital zero (0dBFS). There is usually a indicator red light on the output meter in applications such as Cubase, Logic, Pro Tools, Ableton Live, Fruity Loops , Reaper, Reason, Sequoia. Clipping can be identified after exporting by zooming into the resultant files waveform and checking the peaks of the waveform for "squaring off". Clipping means that the finite amount of numbers in a digital system has been exceeded and therefore the audio signal is no longer accurately represented and is instead just given the last numerical value. Successive clipping points result in a square topped waveform which only drops when the signal level goes below clip point. Re-balancing the mix at this stage is impractical and complex so it is suggested that the master fader be pulled down by the no. of decibels to stop the incidence of clipping. This is a compromise which will have very little effect on the audio quality.

The preferred file formats for mixes are .aiff and .wav.These are both high quality audio file formats which are suitable for mastering.Ideally the files will be at 24bit resolution. (In your digital audio workstation you will find that when you export / bounce your tracks you can choose between 16/24 bit. Choose 24 bit. In addition check that your workstation is set to export "Stereo Interleaved" files. This means that the left and right channel will be locked in time and exported as one single file which is perfect for mastering. The audio sample rate can be left at the same sample rate as the project (i.e. 44.1kHz/ 48kHz/ 88.2kHz/ 96kHz)

It is not ideal to send MP3, Ogg Vorbis or .WMA files. These file formats are lossy formats using (lossy codecs). This means that the data that comprises the music is compressed and some of the data is lost when the audio is data reduced. The algorhythm's employed throw away sonic information which is deemed unnecessary in order to reduce the file size. This means the files are not of the best possible quality and this is not the best for mastering which is trying to maximise your mixes sonic potential.

[Rezzidex]

thanks man this tips really helpd me,,but i wanted to ask something,,i noticed that professional track when i improt them in some wave editing software *i use wave lab,,,) the picture of the wave is all flat and and there is no db that are higher that that flat level...but when i make my track i master it ,,but when i export it it seems that my wave is all messed up .... its similar to this

http://www.google.ch/imgres?imgurl=http ... m=1&itbs=1

,,could you help me with this please

[dubmatters]

Your mix shouldn't be clipping at all. 3-5db of headroom is the minimum you should give the mastering engineer.

Why are you suggesting that:

Re-balancing the mix at this stage is impractical and complex

When it should be mixed properly in the first place.

[eldoogle]

He said to pull down your master fader. Also if your sending to a ME I'm pretty sure you don't dither. The ME will dither.

[dubmatters]

He said to pull down your master fader

You shouldn't need to do that.

[Rubik]

If you are removing a limiter for finalizing purposes you generally WILL have clipping (because the signal is no longer limited to stop at or before 0dB) which in turn will result in you having to pull down the master. I see no flaw in this thought process

[legend4ry]

Adding this to the Production guide.

[-[2]DAY_-]

If you are removing a limiter for finalizing purposes you generally WILL have clipping (because the signal is no longer limited to stop at or before 0dB) which in turn will result in you having to pull down the master. I see no flaw in this thought process

no
nothing oughta be in the red. use leveling amplifiers/limiters on individual channels (think kick, snare)
and nothing on the master.
Granted, the future holds no professional mastering for my tunes, so i compress my master bus (very low ratio) and ride its output fader into a limiter. *shrugs*
I used to over squash my masters and boost them beyond recognition, getting ugly ass mixes as a result
Now i just mix everything hotter and do my little mastering thing on the 2.
but nothings going red, that just aint right.

[hifi]

does bouncing with it interleaved really make a difference to the ME? just wanted to know since I sent some tracks to someone and didn't bounce with that option

[Rubik]

If you are removing a limiter for finalizing purposes you generally WILL have clipping (because the signal is no longer limited to stop at or before 0dB) which in turn will result in you having to pull down the master. I see no flaw in this thought process

no
nothing oughta be in the red. use leveling amplifiers/limiters on individual channels (think kick, snare)
and nothing on the master.
Granted, the future holds no professional mastering for my tunes, so i compress my master bus (very low ratio) and ride its output fader into a limiter. *shrugs*
I used to over squash my masters and boost them beyond recognition, getting ugly ass mixes as a result
Now i just mix everything hotter and do my little mastering thing on the 2.
but nothings going red, that just aint right.

So... You're adamant about having nothing on the master bus but then you both compress and limit the master?I think I may be missing something here

Edit: nevermind dude I think I get you

[gen_]

Why do they actually ask for -3dB when they can pull it down themselves.

[-[2]DAY_-]

So... You're adamant about having nothing on the master bus but then you both compress and limit the master?I think I may be missing something here

yes, this:

the future holds no professional mastering for my tunes, so i compress my master bus (very low ratio) and ride its output fader into a limiter.

lol s'all good

[-[2]DAY_-]

Why do they actually ask for -3dB when they can pull it down themselves.

I am not qualified to explain this, I really don't understand it fully myself, but my intuition tells me quite simply that there is no substitute for actual headroom. Yes, the ME can turn down your track before beginning their work on it, but, that is not only lowering the parts that are too loud. That's lowering the RMS of everything, including the quiet bits that are already sitting where they should. Thus turning those parts way too low in order to get the majority of the stuff down to where it needs to be. I don't necessarily know whether or not this is a problem in every case, but the point therein is:

you ought to get your mix sounding the way you want it to, at a volume level that allows the ME to do their thing. for this to be possible, they need the headroom. for you to get your mix to sound how you want it, you DON'T want to turn around afterwards and lower the master fader because now, in proportion to the loudest parts, the quietest parts are barely audible and won't react well to further processing. (maybe?)


remember peak dB has little to do with how loud something sounds; RMS has much more to do with that. You're better off leveling the peaky transient elements (percussives, snare drums, kicks, etc) in the mix and ride their inputs into the leveler to squeeze out more RMS volume, making your mix sound punchy and loud, but appearing fairly level, and retaining about -3dB headroom.

please correct me if i'm wrong . I'm thinking we just need to read macc's posts again, cuz its all in there.

[knobgoblin]

You really should be sending 24/96 wav files with at least 3db of headroom, preferably 6db. NEVER EVER send any kind of lossy compression file, as mastering will bring the artifacts of this compression to the forefront. take any dynamic and eq processing off the master bus. if you want to send a version of the track with your master bus processing as a guide to the sound that you are after, that would be good as well.

[Ldizzy]

Why do they actually ask for -3dB when they can pull it down themselves.

I am not qualified to explain this, I really don't understand it fully myself, but my intuition tells me quite simply that there is no substitute for actual headroom. Yes, the ME can turn down your track before beginning their work on it, but, that is not only lowering the parts that are too loud. That's lowering the RMS of everything, including the quiet bits that are already sitting where they should. Thus turning those parts way too low in order to get the majority of the stuff down to where it needs to be. I don't necessarily know whether or not this is a problem in every case, but the point therein is:

you ought to get your mix sounding the way you want it to, at a volume level that allows the ME to do their thing. for this to be possible, they need the headroom. for you to get your mix to sound how you want it, you DON'T want to turn around afterwards and lower the master fader because now, in proportion to the loudest parts, the quietest parts are barely audible and won't react well to further processing. (maybe?)


remember peak dB has little to do with how loud something sounds; RMS has much more to do with that. You're better off leveling the peaky transient elements (percussives, snare drums, kicks, etc) in the mix and ride their inputs into the leveler to squeeze out more RMS volume, making your mix sound punchy and loud, but appearing fairly level, and retaining about -3dB headroom.

please correct me if i'm wrong . I'm thinking we just need to read macc's posts again, cuz its all in there.

hum i distantly recall lousy meter resolution being the main reason for the conservative -3 value... as if sometimes, the peak might go above what u believe is zero... but the computer wont even catch it...

but that may be me being wrong..

macc be there please.

[makemerich]

is pulling the master down different than grouping all channels and pulling them down except the master and leaving it at aprox 0?

[daft cunt]

Thanks for that S&S.

One question, there's always a difference between the volume level that Roger Nichols's Inspector is showing in my DAW and the bounced version when I open it in Soundforge.
For the tune I'm working on for instance, I have more than 3dB of headroom but in SF there are peaks at nearly 0 dB.
How is that and which one should I believe ?

[hifi]

thanks for not answering my question i dont mind really. but safe for this guide

[nowaysj]

Thanks for that S&S.

One question, there's always a difference between the volume level that Roger Nichols's Inspector is showing in my DAW and the bounced version when I open it in Soundforge.
For the tune I'm working on for instance, I have more than 3dB of headroom but in SF there are peaks at nearly 0 dB.
How is that and which one should I believe ?

Zoom way in in soundforge, are there actual samples at 0db, or just interpolated curves. Sry, haven't been in sound forge in a lizong time, can't recall how it actually looks.

[serox]

Thanks for that S&S.

One question, there's always a difference between the volume level that Roger Nichols's Inspector is showing in my DAW and the bounced version when I open it in Soundforge.
For the tune I'm working on for instance, I have more than 3dB of headroom but in SF there are peaks at nearly 0 dB.
How is that and which one should I believe ?

Zoom way in in soundforge, are there actual samples at 0db, or just interpolated curves. Sry, haven't been in sound forge in a lizong time, can't recall how it actually looks.

I have noticed this also, what is interpolated curves?!