In this pro mixing tutorial I am going to try to share some of my techniques that I go for when for whatever reason I decided to take on the challenge to mix "crap" (although I am perfectly aware this should not be fixed in mixing or mastering).
Let me just explain what crap typically is in this case. Crap is when you have all of these low quality sound sources jammed inside of that what makes up the recorded stereo signal, could be way too harsh sounding electric guitars, acoustic issues, incredible dynamic issues, noisy background vocals, tons of bad drum bleed with phase issues, weak unbalanced synth signal etc. and that has then been run into a low quality A/D, so not only is it pretty noisy but it also is messed up in its frequency balance, very dead sounding because it does not have any sweet decay and very mono-ish because it is basically too "thin" on each speaker. This is the kind of sound that you typically want to turn off after you've played it for a while. So how can you deal with this.
In a case like this you can do many things, I will exclude the adding sample technique because although that works, that's not what this tutorial is about.
What I do is that I solo each track and then I profile each track, e.g.:
Tight: Yes
Cold: Yes
Organic: Yes
Soft: No
Cool: Yes
Hi-harsh: Yes
Mid-harsh: Yes
Low-harsh: No
...
In this way I am able to create a mix level profile based on understanding the elements that create it and I am now able to isolate the bulk of the noise and better understand how to approach it. Noise is in this case frequencies that don't work, whether they are actually "noisy" in the real sense or not. What I then do is that I apply a temporary master bus fx chain designed to further amplify all of the noise types. You will for instance find filters where I've boosted specific frequencies that I've isolated through frequency sweeping.
Next I listen to the mix in the side component in solo, just to figure out where the bulk of the noise is coming from, I am incredibly focused on getting a soft sounding side component because those frequencies are most close to the listener and must be very sweet for the mix as a whole to be sweet. In the first step I just identify those noise sources. Then I might notice they are fat noise sources recorded in stereo, some of those become mono candidates. Similarly, mono sources that are thin and noisy, those I might make into pseudo stereo, in other words I try to make them add to the stereo image.
Next I might be focusing on background noise, I use certain noise reduction tools for that.
Next I focus on the decay of the tones, because often times you need to add some sustain as well as modulation to these types of sounds to at least make them a bit less dead sounding. Typically I like to run them through 3 very different 200 ms reverbs depending on their panning, separately for center, 50% side and 100% side but on the individual track level, not on the group level. I focus the highest quality reverb on the side panned sound sources that are active on the side component, that makes quite a big difference to the sound as a whole. Typically I avoid pre-delay because that I want to add on a higher more broad group level.
Next I apply zero-phase EQ and do frequency sweeping to isolate and reduce the noisy frequencies on the tracks.
Next I apply zero-phase multi-band compression and do frequency sweeping to isolate the remaining noisy peaks.
I might apply a number of other techniques as well, but at some point when I feel I have dealt with the bulk of the noise, then I begin the process of removing a filter at a time from the master bus. When I've removed all of those filters I am either happy or sad, if I am sad I figure out why, re-enable the filters and resume my work. At some point I am successfully able to remove all of the filters with a good result. At that point what I do is I basically pan the sound sources to be side intensive, I basically let them create as much natural noise as possible, with as realistic/close to final volume balance that I can. It is at that point that I look for mute candidates and just mute those sound sources that add the most noise to the perception of the mix. Once those costly noise sources have been removed, then I dial in the panning in order to create the appropriate amount of bleed into the side component. Now, once I am pleased with the way the side component sounds raw, I balance the rest, this becomes the rough mix. It is at this point that the real mixing starts and I typically apply effects that counter the types of sound qualities I don't like.
At some point I might or might not have the amount of harmonic content present, typically the harmonic content is thin with recordings that have weak signal. In this case it means you need to manually add the harmonic content back in, similarly to how you added decay back in using reverb. In this case you can isolate the dominant frequencies and just add multiples on those on the individual track level, after that you can apply a little pitch shifting and/or chorus and/or tape saturation. A bass for instance might lack a little sub frequencies, maxx bass can help to bring some of that back.
Mixes like this typically require a bit more frequency matching at the end to deal with some of the acoustic issues that the mix to some degree still might reveal. That zero-phase frequency matching is applied on the center panned tracks separately from the side panned tracks. During mastering I might also apply a sound profile that fits well with this type of mix, what that sound profile mostly does is to ensure specific rms levels on specific frequencies, separately on the mid and the side tracks, so let's say I find the mix is a bit noisy in the high end, then I would focus on a sound profile that is extra soft in the high end and frequency match accordingly also. So having the mid panned and the side panned tracks separate is absolutely key to have available for the mastering engineer to work on, especially in a case like this.
During mastering I am also very cautious about how much air exists between each track inside of the mix, because if there is some amount of noise left in there I definitely don't want that noise to be tightly coupled with important sound sources like vocals that then bleeds into the perception of those sound sources, therefore it is important to have the right gain staging in order to get the right amount of air and vibe inside of the mix.
That's basically it. Keep in mind that in many of these cases it might be very important to work with samples, like a really soft kick and snare can work to create better emotions so although there is noise present in the sound it does not mean as much when the mix as a whole feels better. In many cases also there might be really good musical stuff in there too, in that case the noise can become a source of distraction, in those cases just mute it and focus on extracting the value of the really good sounding elements, it could be good rhythm, harmonic chords, good vocal phrases etc. So the more noisy mix is, the more important it is to maximize the quality of the good sounding elements.
Finally, don't over do loudness on these types of mixes, apply a sound profile that makes it big enough but quiet enouth, rms is very important but mixes that have some degree of noise in them should not have individual sound sources competing a lot due to high rms levels, instead focus on ensuring a good groove and bring in the rms carefully with minimum peak harshness.
When you are done with the master, ensure you have several fresh sessions afterwards, so that you can get enough distance to the sound, because sometimes you might get a bit blind to the amount of noise and to the balance quality as well. Also ensure extra good translation, because these types of mixes can become extra noisy on some systems if the mix does not translate extra well.
These were some general ideas on this, I hope this helps.
Let me just explain what crap typically is in this case. Crap is when you have all of these low quality sound sources jammed inside of that what makes up the recorded stereo signal, could be way too harsh sounding electric guitars, acoustic issues, incredible dynamic issues, noisy background vocals, tons of bad drum bleed with phase issues, weak unbalanced synth signal etc. and that has then been run into a low quality A/D, so not only is it pretty noisy but it also is messed up in its frequency balance, very dead sounding because it does not have any sweet decay and very mono-ish because it is basically too "thin" on each speaker. This is the kind of sound that you typically want to turn off after you've played it for a while. So how can you deal with this.
In a case like this you can do many things, I will exclude the adding sample technique because although that works, that's not what this tutorial is about.
What I do is that I solo each track and then I profile each track, e.g.:
Tight: Yes
Cold: Yes
Organic: Yes
Soft: No
Cool: Yes
Hi-harsh: Yes
Mid-harsh: Yes
Low-harsh: No
...
In this way I am able to create a mix level profile based on understanding the elements that create it and I am now able to isolate the bulk of the noise and better understand how to approach it. Noise is in this case frequencies that don't work, whether they are actually "noisy" in the real sense or not. What I then do is that I apply a temporary master bus fx chain designed to further amplify all of the noise types. You will for instance find filters where I've boosted specific frequencies that I've isolated through frequency sweeping.
Next I listen to the mix in the side component in solo, just to figure out where the bulk of the noise is coming from, I am incredibly focused on getting a soft sounding side component because those frequencies are most close to the listener and must be very sweet for the mix as a whole to be sweet. In the first step I just identify those noise sources. Then I might notice they are fat noise sources recorded in stereo, some of those become mono candidates. Similarly, mono sources that are thin and noisy, those I might make into pseudo stereo, in other words I try to make them add to the stereo image.
Next I might be focusing on background noise, I use certain noise reduction tools for that.
Next I focus on the decay of the tones, because often times you need to add some sustain as well as modulation to these types of sounds to at least make them a bit less dead sounding. Typically I like to run them through 3 very different 200 ms reverbs depending on their panning, separately for center, 50% side and 100% side but on the individual track level, not on the group level. I focus the highest quality reverb on the side panned sound sources that are active on the side component, that makes quite a big difference to the sound as a whole. Typically I avoid pre-delay because that I want to add on a higher more broad group level.
Next I apply zero-phase EQ and do frequency sweeping to isolate and reduce the noisy frequencies on the tracks.
Next I apply zero-phase multi-band compression and do frequency sweeping to isolate the remaining noisy peaks.
I might apply a number of other techniques as well, but at some point when I feel I have dealt with the bulk of the noise, then I begin the process of removing a filter at a time from the master bus. When I've removed all of those filters I am either happy or sad, if I am sad I figure out why, re-enable the filters and resume my work. At some point I am successfully able to remove all of the filters with a good result. At that point what I do is I basically pan the sound sources to be side intensive, I basically let them create as much natural noise as possible, with as realistic/close to final volume balance that I can. It is at that point that I look for mute candidates and just mute those sound sources that add the most noise to the perception of the mix. Once those costly noise sources have been removed, then I dial in the panning in order to create the appropriate amount of bleed into the side component. Now, once I am pleased with the way the side component sounds raw, I balance the rest, this becomes the rough mix. It is at this point that the real mixing starts and I typically apply effects that counter the types of sound qualities I don't like.
At some point I might or might not have the amount of harmonic content present, typically the harmonic content is thin with recordings that have weak signal. In this case it means you need to manually add the harmonic content back in, similarly to how you added decay back in using reverb. In this case you can isolate the dominant frequencies and just add multiples on those on the individual track level, after that you can apply a little pitch shifting and/or chorus and/or tape saturation. A bass for instance might lack a little sub frequencies, maxx bass can help to bring some of that back.
Mixes like this typically require a bit more frequency matching at the end to deal with some of the acoustic issues that the mix to some degree still might reveal. That zero-phase frequency matching is applied on the center panned tracks separately from the side panned tracks. During mastering I might also apply a sound profile that fits well with this type of mix, what that sound profile mostly does is to ensure specific rms levels on specific frequencies, separately on the mid and the side tracks, so let's say I find the mix is a bit noisy in the high end, then I would focus on a sound profile that is extra soft in the high end and frequency match accordingly also. So having the mid panned and the side panned tracks separate is absolutely key to have available for the mastering engineer to work on, especially in a case like this.
During mastering I am also very cautious about how much air exists between each track inside of the mix, because if there is some amount of noise left in there I definitely don't want that noise to be tightly coupled with important sound sources like vocals that then bleeds into the perception of those sound sources, therefore it is important to have the right gain staging in order to get the right amount of air and vibe inside of the mix.
That's basically it. Keep in mind that in many of these cases it might be very important to work with samples, like a really soft kick and snare can work to create better emotions so although there is noise present in the sound it does not mean as much when the mix as a whole feels better. In many cases also there might be really good musical stuff in there too, in that case the noise can become a source of distraction, in those cases just mute it and focus on extracting the value of the really good sounding elements, it could be good rhythm, harmonic chords, good vocal phrases etc. So the more noisy mix is, the more important it is to maximize the quality of the good sounding elements.
Finally, don't over do loudness on these types of mixes, apply a sound profile that makes it big enough but quiet enouth, rms is very important but mixes that have some degree of noise in them should not have individual sound sources competing a lot due to high rms levels, instead focus on ensuring a good groove and bring in the rms carefully with minimum peak harshness.
When you are done with the master, ensure you have several fresh sessions afterwards, so that you can get enough distance to the sound, because sometimes you might get a bit blind to the amount of noise and to the balance quality as well. Also ensure extra good translation, because these types of mixes can become extra noisy on some systems if the mix does not translate extra well.
These were some general ideas on this, I hope this helps.
Last edited:
