For all you need to stay home for corona…
I created a large article about the art of mixing :
For all you need to stay home for corona…
I created a large article about the art of mixing :
Sorry about the bad link in the german forum, corrected that.
Wow! Very detailed and comprehensive. Many thanks for sharing.
Thank you! It is a complete book, just saved it to my hard disk for reference!
Yes it was written and kept up to date for a longer time…
I have updated the Basic Mixing tutorial.
Welcome to read the complete mixing tutorial…
Nope, nothing to read or download at the site.
Thank you for sharing this here. I skimmed it over, but will returned to read in detail.
Thank you for an excellent article. There are a few editorial corrections you might want to make to some words and sentences. You use “Base” when you mean “Bass” in almost every instance. Also, in the Mix and Swing section it seems some words are missing from some sentences.
Anyway, again, thank you for writing a very helpful, informative article. I’m still reading it carefully, learning some new things and reviewing basic concepts you cover. Great job, with a few corrections it will be an A+ article.
Thank you, i will correct some words, but it has been an accumulation of me being dutch and spelling conrol all the time.
A table of content would be great. It is interesting that in mixing folks always label the depth dimension ( reverb delay) as dimension 3 and it makes sense mixes are treating the that order! Mathematically ( 3D geometry or Calculus III),
Dim 1 - depth (x)
Dim 2- Panaroma ( y)
Dim 3 - freq. spectrum.
I love the way you have details of almost everything.
I assume your text is written for someone who has not mix at all and want to start. There are terms/techniques such as parallel compression, ducking (in the section: Why does my mix sound so muddy?) that would be nice to have a separate sections for them.
Thanks again for making this available to public.
Thank you to share your article,
there are some good ideas, I liked many explanations and your articles are useful.
There are also questionable ideas about gain staging…To put a limiter on the master bus while mixing is pretty odd. And you suggest -0.3dB, but most authors say that the correct limit is -1dB for the best sound quality. The limiter is used usually in mastering. In mixing it is used in the master track only the compressor to glue the mix instead
You suggest that anything should be under -3 dB. Most contemporary authors suggest that single tracks don’t go above -6dB (some say also less!) in order then when the many tracks are summing up in the master track the overall level should never exceed -1dB and it is very good if the peaks on the master tracks are under-6dB. This produces a better sound quality and help to judge better the song. You miss to explain that 0db digital = -24dB analog. There is not need of a limiter on the master track but only the use of proper gain staging, compression of the tracks and master bus compression. In order to master and give to a mastering engineer the best would be to give him the mix with the maximum peaks no more then -6dB in order there is enough headroom to master. They say that the tracks should have a gain between -24 and -12dB with peaks that don’t exceed -6dB.
It is very questionable also the rule of the need to set the master fader to 0dB. Because internal clipping occurs if the single tracks are clipping, and if peaks of the single tracks of the projects are under -6dB and peaks of the master bus are under 0, then the master fader can be up or down…The dynamic range doesn’t change so much in the digital domain.
This misconceptions leads to the errors to try to reach the O dB while recording tracks on digital systems, with the aim to have more dynamyc range, which is totally wrong. And a mix that on the master bus is always close to 0db, which is not good at all for sound clarity.
Thank anyway for your job which was useful
I am not such an expert but these guys know what they say:
There’s no reason why sound quality should improve when peaks on the master are below -6dBFS as opposed to -1dBFS. None.
Not sure what you mean by that. There is no standard for digital to equal a particular level once converted to analog.
Mastering engineers don’t need peaks to be below -6dBFS for the purpose of having “headroom”. I’m willing to bet that in 99% of all mastering that’s done to at least popular music the track will be made louder, and the range between average and peak will be reduced. Because of this there’s really no need for headroom since generally the difference between peak and average won’t increase.
But even if that was the case the mastering engineer can easily solve any lack of headroom by simply lowering the track before processing. It takes about 1 second to do so and there is virtually zero loss in the process.
thanks for the answer,
the reasons of the equivalence 0dB analog=-24dB digital are explained in detail in the link I posted it is a matter of how the meters are set in the daw and in the analog gear.
About the mix done at -6 -3dB it is an advice found on many sources. If all that the mastering engineer has to do is to lower the volume of the mix, I don’t understand why mastering engineers complain so much about the lack of headroom in mixes they receive, and even refuse mixes for that!
it happens that even if in our DAW there is not clipping and nothing goes above 0, during the digital to analog conversion process there is clipping (inter sample peaks) …
You are right there is not better sound quality in -6dB then -1dB. I was wrong and I didn’t explained correctly my point. When you pass -1dB we enter the dangerous zone of inter sample peaks. then leaving more headroom it is just to leave a bigger safety zone from the dangerous zone
and experience teaches that when I aim for -3db I often end up louder…If then you realize you are too loud, there is no more room for EQ or to raise the volume of a track, a vocal for example. And you have during mixing to lower the volume of all your channels…it is not 1 second!
This is why the advice to keep the levels low makes sense to me. There is no need to go close to 0 and there is no need to make the mastering engineer to waste a second of his time.
in the site of Waves plugins they have this on mastering:
"Start with Sufficient Headroom
The second golden rule – actually, a corollary to the first one – is: Always leave yourself sufficient headroom. Most plugins – especially those that model analog equipment – are designed to work at around a particular input level range, and that range definitely does not include peaks at zero. A good rule of thumb is to keep steady signals such as rhythm guitars, synths or pads at somewhere between -20 and -16 dBFS, with transient peaks (such as occur from drums and percussive instruments) no higher than -6 dBFS."
Then summing up: approaching the -1dB 0dB we are approaching the “limit” zone, where things mess up. Plugins are on their stress zone, inter sample peaks may occur, artifacts and distortion are more likely to happen when we export the mix… Have you ever heard about loudness war and how today everything is too loud and compressed and squashed with bad dynamics, just to approach 0dB…both in cinema an music?https://www.soundonsound.com/techniques/end-loudness-war
Anyway we know…there are many different ways of thinking about mixing and mastering
Well, you did write the opposite before “0db digital = -24dB analog” rather than “0dB analog=-24dB digital”.
Also, like I said, the nominal reference level (0VU) on a converter can equal anywhere from -16dBFS to -22dBFS. So there is no standard. The devices are calibrated differently.
In that article on gain-staging it seems to me that the author is using “dynamic range” and “headroom” interchangeably, and that’s not really right I think. I think most professional engineers would agree that having too little dynamic range (if we assume that means the range between peak and average) is bad when sending stuff out to mastering. But that’s different from headroom. The space between nominal operating level and the onset of distortion in analog (i.e. 0VU to distortion) is the headroom. It’s ‘above’ something. So leaving space above the peaks (to 0dBFS) in a mix doesn’t really do much since a mastering engineer can just lower the mix easily and is likely - as suggested - to diminish the range rather than extend it.
I just don’t see the point in leaving for example 6dB of space as a general rule.
Yes, I have heard about the loudness war… It has very little to do with “-1dB” or “0dB” and everything to do with adjusting dynamics and increasing loudness.
Thanks for offering constructive comments here.
I’m not sure if the OP covers Crest Factor in the article? I began reading it but there were too many typos and some apparent translation issues making it impossible for me to continue. Below is a link to one article on the topic for anyone’s reference, but I’d be curious to hear anything you might have to say about it. The article is from 2013 so is perhaps a bit dated in some respects.
Hope all’s well.
I think the article is maybe dated when it talks about access to metering since I feel we have access to pretty much anything today, but other than that I agree with it and makes the same point I made. The loudness war isn’t about getting close to 0dBFS but what to do with “the rest” of the signal.
I don’t think the OP covers ‘crest factor’ since doing a search didn’t find anything. I didn’t read the whole thing though since I didn’t care for the formatting, and quite frankly I’ve done this for long enough not to have time to read notes on the basics of mixing.
Since I searched for the term “average” I found the section on “Loudness Maximizer Effects.”, and I don’t think that it’s correct to say that it is “A combined effect of gaining and compression (limiting) for the purpose of getting the most dynamic out of a mix”. “most dynamic” to me implies more dynamics, and that to me is the opposite of a louder mix. Louder mixes have less dynamics the way we think of the word “loud”.
The very first sentence on iZotope’s page on their loudness maximizer reads: “The Loudness Maximizer allows you to create an overall louder or fuller master by limiting the dynamic range”
I think if I read the entire article I might find a few of those cases where the terminology isn’t really used correctly. Could be a language barrier or whatever.