MusicWater
Banned
In terms of mixing I find that there are several "pitfalls" that tend to cause weak mixes. One is to have the wrong volume/compression ratio and too loud volume on heavy duty sound sources such as vocals. For me it has varied what ratio I find most optimal, but currently I find vocals should be heavily limited and then lowered in volume as much as possible. What seems to happen when vocals are left too dynamic in the mix is that you automatically compensate with volume boost, this in turn becomes very expensive on the mix signal, so that you don't get enough signal available to all other sound sources in the mix, especially not in the chorus. To avoid this situation I usually start mixing without vocals.
Early on I mix the kick drum, snare and bass as if they were the only sound sources in the mix. I want these to be fat, big, powerful but absolutely not too "heavy". A lot of focus on compression and EQ and using the right speakers for the job. Once this group is balanced I add vocals into the mix, the focus is on making the vocals force the low end mix signal down as little as possible, so that as much of the low end fatness as possible is preserved and so that I have some room left for other sound sources too, typically that means peak limiting rather than rms limiting. Once these two are unified, then in order to fit the other sound sources into the mix, I lower both the vocal mix and the low end mix in equal doses, however as little as possible.
Once all of these sound sources have been packaged into the mix signal like this, I usually end up with a very fat mix. If I do it some other way, let's say I don't add any compression and start with vocals instead and do free mixing, what happens is that the mix signal is lost to various dynamic sound sources that are set loud in the mix, causing a situation when it's dynamic but very weak and non-powerful and it does not sound good, or it can sound good but not powerful and great.
During mastering I want to be able to kind of take the mix and just enlarge it as a rectangle, with mixes that are very unbalanced in how they consume the mix signal (wrong types of compression/limiting on different sound sources), it becomes tricky to enlarge the mix as a rectangle, what you instead tend to end up with at higher rms levels is an overall "hard" and unbalanced mix that lacks good stereo qualities. Once I have succeeded with the mix balance, in mastering I force the loudness to "meaty" levels. Early on I don't care about clipping, all I care about is finding that meaty spot where the whole mix comes alive in a really major way. That's the most important thing, that's where it needs to be. Usually this means a lot of gaining and re-balancing in iterations until the mix is "meaty". Once it sits there, I leave it there even when I have clipping issues in the mix. The clipping I take care of later. I could have chosen a different route here too, for instance focused on the clipping. That gives you a mix that might be loud but not meaty.
I find it's much easier to remove clipping noise than to get stuck with a mix signal that simply always lacks meat. How I work with this in mastering is that I lock down the mix down to song sections and make each song section have that meat, in other words I don't focus on being aware of what should happen about the volume at various song parts, such as chorus being louder than verse or focus on the loudness of the mix as a whole, instead I make each song section having its own "best meat". Once this is done, then I automate the signal level based on the goals of how it should progress. But this makes the overall meat about the mix as a whole come up to a commercial level.
Early on I mix the kick drum, snare and bass as if they were the only sound sources in the mix. I want these to be fat, big, powerful but absolutely not too "heavy". A lot of focus on compression and EQ and using the right speakers for the job. Once this group is balanced I add vocals into the mix, the focus is on making the vocals force the low end mix signal down as little as possible, so that as much of the low end fatness as possible is preserved and so that I have some room left for other sound sources too, typically that means peak limiting rather than rms limiting. Once these two are unified, then in order to fit the other sound sources into the mix, I lower both the vocal mix and the low end mix in equal doses, however as little as possible.
Once all of these sound sources have been packaged into the mix signal like this, I usually end up with a very fat mix. If I do it some other way, let's say I don't add any compression and start with vocals instead and do free mixing, what happens is that the mix signal is lost to various dynamic sound sources that are set loud in the mix, causing a situation when it's dynamic but very weak and non-powerful and it does not sound good, or it can sound good but not powerful and great.
During mastering I want to be able to kind of take the mix and just enlarge it as a rectangle, with mixes that are very unbalanced in how they consume the mix signal (wrong types of compression/limiting on different sound sources), it becomes tricky to enlarge the mix as a rectangle, what you instead tend to end up with at higher rms levels is an overall "hard" and unbalanced mix that lacks good stereo qualities. Once I have succeeded with the mix balance, in mastering I force the loudness to "meaty" levels. Early on I don't care about clipping, all I care about is finding that meaty spot where the whole mix comes alive in a really major way. That's the most important thing, that's where it needs to be. Usually this means a lot of gaining and re-balancing in iterations until the mix is "meaty". Once it sits there, I leave it there even when I have clipping issues in the mix. The clipping I take care of later. I could have chosen a different route here too, for instance focused on the clipping. That gives you a mix that might be loud but not meaty.
I find it's much easier to remove clipping noise than to get stuck with a mix signal that simply always lacks meat. How I work with this in mastering is that I lock down the mix down to song sections and make each song section have that meat, in other words I don't focus on being aware of what should happen about the volume at various song parts, such as chorus being louder than verse or focus on the loudness of the mix as a whole, instead I make each song section having its own "best meat". Once this is done, then I automate the signal level based on the goals of how it should progress. But this makes the overall meat about the mix as a whole come up to a commercial level.
Last edited: