Practical Mixing

Why your mix sounds bad in the car (and what the car is actually doing)

By StudioRescue Editorial

Published October 8, 2026

The car test is a genuinely useful habit. It is also the most commonly misread listening test in home mixing, because people treat the car as a neutral second opinion when it is nothing of the kind. A car is a small, oddly shaped, loud room with an opinionated stereo in it. Every one of those things changes what you hear, and none of them is in your project file.

That matters because the usual response is to go back to REAPER and start cutting bass. If the low end was never actually too loud, you have just made the mix thin everywhere else to please one vehicle. So before changing anything, it is worth knowing what the car is contributing.

What the car is doing before you blame the mix

1. The cabin’s low end is a few big resonances, not a response curve

A car interior is small enough that its lowest acoustic modes land right on top of the bass range. In one published measurement of a coupe interior, the first longitudinal and lateral modes sat at 64, 73, 108 and 115 Hz. Below roughly 300 Hz there are only a few widely separated modes, which is a very different situation from a response you could draw as a smooth line.

Two consequences follow, and both are practical. Low-end behaviour in a car depends heavily on where you are sitting, because you are sitting inside a standing-wave pattern. And it varies a lot between vehicles, because mode frequencies depend on the shape and size of the cabin. A sedan, a coupe and an SUV do not agree with each other. That is why there is no such thing as mixing for “the car.”

2. Road noise sits exactly where your mix is most vulnerable

Road noise in a modern vehicle typically occupies about 20 Hz to 1 kHz, with most tire and road interaction noise concentrated between 100 Hz and 500 Hz. Below about 300 Hz it is mostly structure-borne, arriving through the suspension and body rather than through the air, and tire cavity resonances add tonal content around 200 to 250 Hz. Low-frequency components usually dominate what you notice in the cabin.

Read that range again, because it is the surprise. Road noise is not a hiss sitting on top of your mix. It is a loud, uneven blanket across the bass fundamentals and the low mids, which is where a kick, a bass, and the body and weight of a lead vocal all live. At speed, that content is partially masked. Your mix did not lose its vocal; the vocal is being covered up in the band where it carries weight.

This is why the common reaction of brightening the vocal is usually wrong. Masking at 200 Hz is not fixed by adding 3 kHz. It makes the vocal harsher without making it any more solid.

3. The stereo is changing your mix while you drive

Many cars actively process audio in response to speed, and the manufacturers that do document it plainly in their own owner’s manuals. Mazda’s ALC “automatically adjusts audio volume and sound quality according to the vehicle speed.” Bose AudioPilot “continuously adjusts the music to compensate for background noise and vehicle speed.” Volvo’s volume compensation “compensates for disrupting noises by increasing the volume in relation to the speed of the car.”

Note that Mazda’s wording says sound quality, not only volume. A system that changes tone as a function of speed is not a reference. Whatever you concluded at 30 mph may not be what the system was doing at 65 mph.

4. You are probably not listening to your file

If the audio reached the car over Bluetooth from a streaming app, it was loudness normalised on the way. Spotify states that it adjusts tracks to -14 dB LUFS according to the ITU 1770 standard, with selectable playback levels of -11, -14 and -19 dB LUFS. A quieter master gets positive gain and a louder one gets negative gain, so two versions of the same mix can arrive at two different levels.

Level is not a detail here. Perceived tonal balance changes with absolute playback level, which is the same reason an unmatched reference track misleads you. That mechanism is covered in detail in how to use a reference track in REAPER and level-match it first, and it applies just as much to a car as to a pair of monitors.

A car test that actually tells you something

The goal is not to hear your mix in a car. It is to separate the four effects above so that what is left is attributable to the mix.

What to change in REAPER, and what to leave alone

When the car test does point at the mix, it is usually one of a small number of things. ReaEQ ships with REAPER and offers low shelf, high shelf, band, high-pass and notch shapes, with frequency, gain and bandwidth per band, which covers everything below. Treat these as starting ranges to test, not settings to apply.

For the masking row, the fix is competition, not brightness. If the vocal and the bass instruments are both occupying 150 to 400 Hz, road noise takes out that whole region and the vocal loses its body first. Clearing low-mid buildup on the vocal so it is not relying on a crowded band is covered in why your home vocal sounds muddy in REAPER, and deciding how far up to push it afterwards is covered in how loud vocals should be in the mix.

For the last row, a mono check is worth doing before you touch EQ at all, since anything that cancels will also thin out on phones and laptop speakers. The mono-compatibility part of that is covered in how to make vocals wider in REAPER without breaking mono.

What this will not fix

• It will not make your mix sound the same in every car. Cabin modes differ by vehicle, so a mix tuned to one car is mistuned for the next. Translation means acceptable everywhere, never identical anywhere.
• It will not survive the car’s own processing. If the driver leaves speed-dependent volume and a bass boost switched on, the system will reshape your work regardless of what you did.
• It will not substitute for the room. If your mixes are consistently wrong in the car, the likeliest single cause is that you cannot hear the low end where you work. A car test diagnoses that; it does not solve it.
• It will not recover masked detail. At 65 mph some of your mix is simply inaudible, and no amount of EQ changes the noise in the cabin. Mixing so the important things sit above it is the realistic goal.
• It is not a loudness test. What reaches the car has usually been normalised, so the car cannot tell you whether your master is competitive.

If it still does not add up

1. Check you really disabled the car’s compensation. Some systems re-enable it per source or per profile. If the volume changes when you accelerate and you did not touch the dial, it is still on.
2. Check you are not listening to a stream. Bluetooth from a phone app is the default on most cars and the easiest thing to forget.
3. Level-match before judging. If the car playback is louder than your monitoring was, it will sound fuller for that reason alone.
4. Try a second car, or any unrelated system. One car is a sample of one. Two disagreeing cars tell you more than one confident one.
5. Suspect the room before the mix. A mix that is wrong in the same direction every time, in every car, is usually a monitoring problem rather than a judgement problem.