MIXING AND MASTERING

[magnesson]

i don't want to sound like a noob but i can not mix and master i don't know why but i can't i tend to do less musically to keep from getting muddy. which generally happens anyways with all the experimentation i do. so my question is, is there any one in our ambient family who's bread and butter is mixing and mastering? i am trying to do a physical release but it sounds like a 90's mix tape comp. any suggestions??? much appreciated!!

-magnesson

[aoVI]

... i am trying to do a physical release but it sounds like a 90's mix tape comp. ...

That's what I shoot for every time.

But seriously, someone should have some good suggestions soon.

[synkrotron]

It's all about watching those frequencies, both on the individual tracks and the master bus. I stick an instance of Pro-Q2 on every track, before I even start.

Can't say I know exactly what I am doing, but when I listen and analyse some of my older stuff I can honestly say that the stuff I have been producing over the last two years is much, much better.

I have bookmarked the following websites and refer to them often:-

http://thehub.musiciansfriend.com/tech-tips/eq-curves-and-musical-style

http://soundbytesmag.net/spectraldynamicsdynamiceq/

https://www.residentadvisor.net/features/1895

http://www.teachmeaudio.com/

http://www.independentrecording.net/irn/resources/freqchart/main_display.htm

Some of the above are specific to a particular tool but they are still worth a read

[Organic Patterns]

I agree with what Synkrotron says, put a EQ on each individual track and the master bus (keep this one untouched, it's only for analysing the frequency range of the entire project). If you look at rock music for instance, each instrument has its own place in the frequency range, it's therefore relatively straight forward. This doesn't come as natural in ambient, or other electronic genres, so you need to be more aware of designing for each frequency range yourself. When you start the sound design for a track, think about its frequency range in comparison to your other tracks. Too many tracks within the same frequency range will make it muddy. Another thing to keep in mind is, each track usually has a part of its frequency range which is dominant. It might stretch from low to high, but ask yourself which part within its frequency range is dominant. Those parts which aren't dominant, might not even be needed, and in some cases you might not even hear them, therefore it could be beneficial to cut them out.

What I usually do when I have to many tracks within the same frequency range where I don't wish to cut any frequencies out, is I automate the volume, so that each track is at its highest volume at its own time. Say I have 4 interfering tracks in 32 bars, my first track is peaking at the first 8 bars, the second track is peaking at the next 8 bars and so on. At the 8 bars up till the peak of a specific track, and the 8 bars after the peak I would fade them in and out like this:
Screen Shot 2016-05-02 at 14.01.32.png
By doing this I have 3 tracks playing at all times, a track at its highest and two tracks fading either in or out. This allows me to maintain the sounds of all 4 tracks, and it creates a nice floating progression. These 4 tracks playing at its highest volume at all times would make it muddy, but by automating it they peak at its their own time, and problem would be solved.

Well that's my two cents, hope it helps :-)

[magnesson]

I agree with what Synkrotron says, put a EQ on each individual track and the master bus (keep this one untouched, it's only for analysing the frequency range of the entire project). If you look at rock music for instance, each instrument has its own place in the frequency range, it's therefore relatively straight forward. This doesn't come as natural in ambient, or other electronic genres, so you need to be more aware of designing for each frequency range yourself. When you start the sound design for a track, think about its frequency range in comparison to your other tracks. Too many tracks within the same frequency range will make it muddy. Another thing to keep in mind is, each track usually has a part of its frequency range which is dominant. It might stretch from low to high, but ask yourself which part within its frequency range is dominant. Those parts which aren't dominant, might not even be needed, and in some cases you might not even hear them, therefore it could be beneficial to cut them out.

What I usually do when I have to many tracks within the same frequency range where I don't wish to cut any frequencies out, is I automate the volume, so that each track is at its highest volume at its own time. Say I have 4 interfering tracks in 32 bars, my first track is peaking at the first 8 bars, the second track is peaking at the next 8 bars and so on. At the 8 bars up till the peak of a specific track, and the 8 bars after the peak I would fade them in and out like this:
[ATTACH=CONFIG]1374[/ATTACH]
By doing this I have 3 tracks playing at all times, a track at its highest and two tracks fading either in or out. This allows me to maintain the sounds of all 4 tracks, and it creates a nice floating progression. These 4 tracks playing at its highest volume at all times would make it muddy, but by automating it they peak at its their own time, and problem would be solved.

Well that's my two cents, hope it helps :-)

this makes a lot of sense thank you!

[Organic Patterns]

I'm happy to help! :-)

[ian]

as a general mixing tip, high-pass anything lower than 20Hz on the master. these frequencies are inaudible and deprive you of headroom. in ambient stuff it's okay to high-pass more aggressively; if you're not using sub bass in your mix then you can even get away with putting the cutoff anywhere between 60-90 Hz, or even potentially greater depending on how much the mix needs bass frequencies.

similarly, low-pass your highs as aggressively as possible on every mixer track. this is to avoid aliasing, which can be caused by any multitude of plugins and also deprives you of headroom. i made the default preset of my equalizer cut off everything higher than 17kHz since that's beyond the hearing range of the vast majority of the listening audience. it's pretty much standard practice to cut anything higher than 17kHz on the master.

reverb can sound great on pads, drones, gated stuff, and a multitude of other ambient mainstays. however, it's easy to over-do. as a general rule of thumb, the more uninterrupted frequency movement an element in your track has, the less reverb it needs. overdoing reverb can make stuff sound muddy because it increases stereo presence and blurs different parts of a sound together. reverb can be especially mud-inducing on bass frequencies because of this; if i want reverb on a bass i like to bandsplit it so it only affects the higher frequencies of the sound and/or minimize the stereo presence of the reverb and/or use a short decay time/low mix level/etc.

over-using compressors can make stuff sound muddy due to reduction of dynamic range and compressors' tendency to mess with time variables of sounds. i only ever use a compressor on the basses bus and master channel because of this. i put one on the bus all my basses go to because i like my basses to play at a more consistent level and sound "punchier", and i put one on the master because mine doubles as a limiter and it's often desirable to have the track be a few dB louder/"filled out" more/the effect is much less noticeable if it's applied to the whole track. if i wanna make an instrument louder i tend to just crank up the volume on the mixer or vst itself, or use a vst dedicated to adding volume to a sound.

keep an eye on sounds that peak higher than 0 dB; this is important for 24bit or lower playback (what the entirety of your listening audience will hear). if a portion of the track peaks higher than 0 dB and you don't have a limiter on the master then your audio clips. if you do have a limiter, the sound becomes unavoidably compressed and you get deprived of transient information. this isn't always a bad or even noticeable thing but it can sound like horseshit if there's too much of it in the wrong places. probably not as realistic an issue in more minimal and quiet ambient stuff, but it's still something to look out for anyway.

this can be really minor but watch your plugin delay compensation. certain samples and vsts add a few milliseconds of latency to mixer channels, which can make even totally dry and gridlocked sounds overlap. in more busy stuff i like to just set this manually or automatically trim pdc when i export the project to an audio file.

a lot of the dirt in basses comes from the 100-500Hz range. if you find that your basses sound muddy then use your ears to appropriately attenuate these frequencies.

if you're using drums then sidechaining most of your track to them is almost always a good idea. the principal drums (kick and snare) pretty much always need to be louder than everything surrounding them so as to simulate realism (go to a live band's performance; when are the drums ever quieter than other elements in the song?) and provide a better sense of groove. sidechaining is a good way to do this. if you find your drums sound good without sidechaining then it's because the music around them is gain-staged appropriately.

use delay smartly. more slap-back style effects are cool on sounds with long decay times but longer delay can add unnecessary sludge to them. long delay can sound cool on sounds where the transient is the main focus, but overdoing this or applying it to too many sounds in the same frequency range muddies a track in much the same way that reverb does.

that's all i can think of for now. pretty much all of this applies to genres outside the realm of ambient music, too.