Guides · Referencing
A/B your mix against a reference
Comparing your mix with a finished record you trust is one of the fastest ways to hear what's off. It only works if the comparison is fair, and that mostly means matching the levels.
Why reference at all
Your ears adapt. After an hour on one song, a slightly dull top end or a big low-mid bump starts to sound normal. A reference track resets you: a few seconds of a record you know well and you immediately hear that your mix is darker, wider, or has less low end than you thought.
Referencing is also how you calibrate a new room or new headphones. You know how that record should sound, so any difference is your monitoring.
Level-match first
The biggest trap in any A/B comparison is level. Given two versions of the same thing, most listeners prefer the louder one, even by a small amount. A commercial master is usually louder than a mix in progress, so an unmatched comparison will always make your mix sound worse, and you'll start fixing the wrong things.
- Turn the reference down, not your mix up. Your mix bus should stay where it is.
- Match similar sections. Compare the chorus of your mix with a chorus of the reference, not an intro.
- Use loudness readings as a start. See LUFS explained; short-term loudness over the same kind of section is a good guide.
- Finish by ear. When switching makes neither sound obviously "bigger", you're close.
Choosing references
- Same genre and density. A sparse acoustic song isn't a useful reference for a dense rock mix.
- A version you trust. Use a high-quality file, not a low-bitrate stream or a video rip.
- Songs you know on many systems. If you've heard it in cars, clubs and earbuds, it's a better reference on each of those.
What to compare
Switch quickly and focus on one thing at a time. Short switches of a few seconds work better than long listens.
- Tonal balance. Low end, low mids, presence, air. Which way does your mix lean?
- Vocal level and tone. Where does the lead sit relative to the band?
- Low end. How do the kick and bass share the bottom? How much sub is there?
- Width and depth. How wide is it? How much reverb? Try it in Side too.
- Dynamics. How much does the chorus lift from the verse?
Reference, don't copy
The goal isn't to make your mix identical to someone else's record. It's to notice where yours differs and decide whether you meant it.
Other things worth A/B-ing
- Two mix versions. Last night's bounce against today's session.
- A bus against the full mix. A vocal bus or drum bus in context, and on its own.
- Before and after the mix bus chain. Is the bus compressor and limiter actually helping at matched level?
Compare in MixOutpost
MixOutpost streams your mix from the Mac to your iPhone, and Compare adds a second source as B. You can then switch between A, B and A+B on the phone. Switching is ramped, so it doesn't click, and it works on any output: the phone speaker, AirPods or AirPlay.
- 01
Set up a second source
Any other MixOutpost source on the same Mac can be B. Pairing covers every source on that Mac, so there's nothing extra to pair.
- 02
Add it as B
In Sources, tap Compare next to the second source. It becomes B and the compare bar appears in Listen.
- 03
Match levels, then switch
Each source keeps its own volume, from −40 to 0 dB, remembered per source. Turn the louder one (usually the reference) down until they match, then switch A / B / A+B.
Listen as applies to what you're hearing, so you can compare A and B in Mono or Side too.
Two ways to get a reference in as B
A second plugin instance
Put the reference track on its own track in the DAW, routed to an output that doesn't feed your master, and insert a second MixOutpost instance on it. It shows up on the phone as its own source ("Host · Track"). The reference plays in sync with the session, and your master bus stays clean.
The MixOutpost Mac app
The Mac app can share Mac audio (all apps), so you can play a reference from a music player and add the app as B. Note that it captures everything the Mac plays, including your DAW, so B is cleanest when the DAW is stopped. It can also share an audio input device instead.
Compare is one of the features proposed for MixOutpost Pro, a one-time in-app purchase. Several people can also listen on their own phones at once; see multiple listeners.
Questions
Why does level matching matter so much?
Because louder almost always sounds better at first: fuller, brighter, more exciting. If the reference is a few dB louder than your mix, you'll hear that as a quality difference. Matching levels removes the bias so you hear the real differences.
Should I match by LUFS or by ear?
Start with loudness measurement if you can, then fine-tune by ear on the section you're comparing. Integrated LUFS over a whole song is a good baseline; short-term loudness on the same section is closer to what you hear in the moment.
How many reference tracks should I use?
Two or three is plenty: songs in the same genre and arrangement density, released in a format you trust. More than that and you spend the session listening instead of mixing.
What is A+B for?
Playing both at once is useful for checking timing and alignment, for example when comparing two versions of the same mix or a stem against the full mix. For judging a reference track, switching between A and B is more useful.
Hear your mix on your phone
Free Mac plugin (AU, VST3) and iPhone app. Lossless, bit-exact and encrypted.