Stems work well if you’re sending the files to a vocalist to record to. Rather than sending up to 100+ individual instrument tracks, you can simply send a handful of stems-one for each major instrument group-to simplify things ahead of time. Unless your collaborator has specifically requested you send them a multitrack, stems are probably the best way to go. It’s always best practice to export your multitracks as “dry” as possible-meaning you should only include effects that are crucial to the sound of the instrument, but nothing more. For this reason, multitracks typically exclude any use of compressors, delays, or reverbs so that these decisions can be left up to the mix engineer. For example, in order to get everything to sit well together in a mix, a mix engineer is going to need absolute control over each individual instrument. Multitracks are typically requested when complete flexibility is needed. Whereas the track counts in multitrack sessions range from twenty to a couple hundred, stem sessions may contain only four to twenty tracks.īoth stems and multitracks can be useful, depending on your desired outcome, but sometimes sending multitracks is overkill, and sometimes sending stems just isn’t enough. Helps isolate the exact stage of the process in which any particular problem emerges, making the diagnosis and resolution of said problem FAR simpler.As you can see, multitrack sessions typically have far more tracks than stem sessions. Prevents backtracking / wasting time by compartmentalizing the processįrees up system resources, resulting in cleaner sounding stems

TL DR In summary, stemming and splitting your process up into stages is useful for several reasons: With multi-track releases like EPs, I've started using an 'Overlord' Mixing & Mastering template for all of the tracks in the release to make sure they all have a consistent sound. All in all, I usually only end up with a max of 3 project files per track. Then, if I'm ever tempted to change something from an earlier stage, the somewhat annoying process of unfreezing & re-freezing the channel stands in my way to ensure that I only actually do it if it REALLY needs to be done.įinally, in the Mixing stage, I add EQ, panning, and levels automation to sort out any remaining headroom issues before heading into the Mastering phase. I find that this process turns out much crisper, tighter, and more consistent sounding stems than if I was processing all of the VSTs/AUs/Resamples/etc alongside their respective effects chains. Then, when the Arrangement stage is complete, I freeze the existing channels, bounce their stems into new channels inside the same project file, and then begin adding effects to them. During the early stages, I try to focus strictly on the raw sounds, and ignore the urge to polish them with effects and EQ. In doing so, I've given each stage of the process a bit more isolation and sovereignty, and prevented myself from regressing to earlier stages when I should be focusing on finishing up the track.Įven more recently, I've been experimenting with adding an Effects stage between Arrangement and Mixing, and this has made me appreciate the stemming process even more. Now, I've broken up my process into stages (Sound Design > Composition > Arrangement > Mixing > Mastering) and started using different project files to establish a more tangible distinction between them. Until recently, I always found myself going back and re-tweaking stuff that I'd already done quite well, and just wasting time fiddling with stuff that you'd never even be able to hear in the final mix anyway. If you're doing your own mastering, then stemming out is a fantastic way for you to break the production process down into its individual 'stages' and force yourself to work on one at a time.

It allows for greater control over the end-product. In conclusion, stem mastering is just another option.

Stem mastering for this purpose would involve a subgroup your whole song except rhythm, then a subgroup of your rhythm section without kick, then a stem containing only your kick drum. To get max RMS, you typically want to mix your kick drum a lot louder than you normally would, and then limit the hell out of it and bring the rest of the track up to the kick's level with your master chain. You can more easily and accurately compress separate groups of instruments (instead of just relying on multiband compression), and it's easier to say, make your leads wider without changing how the drums sound.Īnother reason to stem master is for maximum RMS. The main reason to stem master is that you get a bit more control over the final master. All of your basses will be on one, your drums on another, your leads on another, etc. There may be 3-8 stems for stem mastering. Stem mastering doesn't involve every stem of a track, but rather subgroups. Stem mastering is essentially the same as stereo mastering.

Stereo master vs stem master