Linkwitz-Riley Crossovers: Why Your Bass Got Tighter
We've enabled proper 4th-order Linkwitz-Riley crossovers in the multiband processing of the mastering engine. This change is audible — so here's an honest breakdown, with numbers, of what changed and how we verified it.
The problem: band boundaries were eating bass
Multiband dynamics splits a track into frequency bands, processes each independently, and sums them back. For the sum to be transparent, the bands must meet perfectly — that's the crossover's job.
A DSP-engine audit showed our previous implementation (double Butterworth filtering) did not produce a correct LR4 sum. Energy was lost where bands meet, and the low end suffered most: on EDM material the bass deviation from the reference spectrum reached −4.15 dB.
The fix and the verification
The corrected crossover is a classic 4th-order Linkwitz-Riley — the mastering industry standard: bands sum to a flat response with no dips at the crossover frequencies.
We didn't enable the fix by ear alone. Test tracks in three genres were rendered by the old (A) and new (B) implementation, and both were compared against a data-derived reference spectrum curve of commercial releases (based on AES research):
| Genre | Bass deviation, A → B | Integrated spectrum error |
|---|---|---|
| EDM | −4.15 → +0.10 dB | 0.194 → 0.035 |
| Hip-hop | unchanged | 0.933 → 0.857 |
| Acoustic | −1.95 → −1.44 dB | 0.728 → 0.704 |
The new implementation is closer to the reference in all three genres. EDM bass landed almost perfectly — the old crossover really was eating it.
What it means for you
- New masters get a more accurate, tighter low end — especially EDM, hip-hop and bass music.
- Target LUFS and true peak are guaranteed exactly as before — only the tonal balance at band boundaries changed.
- Already-downloaded files don't change. Want to compare? Re-master a track and A/B it in the app.
Mastering quality is the core of our product, so every change to the sound is verified with measurements and kept revertable. If any track seems to sound worse — hit 👎 in the app: we investigate every such case on the actual file.
Try it: magicmaster.pro/app?lang=en — 1 free basic mastering per UTC day.
Questions and answers
What is a Linkwitz-Riley crossover and why is it the standard?
A crossover splits the signal into frequency bands for independent processing (multiband compression, limiting). 4th-order Linkwitz-Riley is the industry standard: its bands sum back to a perfectly flat response, with no dips or bumps at the crossover frequencies.
What was wrong with the old implementation?
The double filtering didn't produce a correct LR4 sum: energy was lost where bands meet, most of all in the bass. On EDM material the bass deviation from the reference spectrum reached −4.15 dB.
Will my old masters change?
Files you've already downloaded don't change. New masters get a more accurate low end: measurements show the balance is now closer to the reference spectrum of commercial releases in every genre we tested.
How did you verify it got better, not just different?
With objective A/B measurements: the same tracks rendered by the old and the new implementation were compared against a data-derived reference spectrum curve of commercial releases (based on AES research). The new version is closer to the reference in all three test genres — EDM, hip-hop, acoustic.
Upload a track — a finished master in seconds.
Open mastering →AI mastering onlineLUFS analyzer