What Mastering Actually Changes — With Numbers
You ran a remaster, pressed play, and thought: it sounds… the same? That's the most common reaction to a properly done master — and it's also the biggest misunderstanding about what post-production is for.
A good master rarely sounds like a different song. It sounds like the same song with the friction removed. The changes are real and measurable — you just need to know where to look and what to listen for.
The Numbers That Actually Moved
Every KROMA result reports two measurements: integrated LUFS (perceived loudness) and true peak (the highest instantaneous level). A typical remaster might read Original −16.4 → Processed −10.2 LUFS. That 6 dB of headroom gain means your track sits at streaming-competitive loudness without clipping.
Why this matters: every major platform normalizes loudness. Spotify targets roughly −14 LUFS, YouTube similar, Apple Music −16. If your track arrives at −22 LUFS, the platform plays it quieter than everything around it — listeners reach for the volume and subconsciously rate your song as weaker. If it arrives crushed at −6 LUFS, the platform turns it down and the distortion stays.
How to Actually Compare
Don't listen for loudness — louder always sounds 'better' to human ears, which is exactly why engineers match levels when they A/B. Listen for texture instead:
• High end: do cymbals and vocal air have more space around them, or less? • Low end: does the kick land tighter, with less rumble underneath? • Vocals: do consonants (s, t, k) sit more clearly on top of the backing? • Transitions: do quiet verses and loud choruses feel more intentional?
Toggle the ORIGINAL button while the track plays — the playhead stays synced, so you hear the same second of music in both versions. That instant A/B is where the difference stops being subtle.
What 'Good' Looks Like
For streaming releases, aim for −9 to −14 LUFS integrated with true peak at or below −1.0 dBTP. For video uploads, slightly quieter is fine — platforms normalize anyway. If your processed result lands in that window with no red peaks, the master did its job even if it didn't sound dramatic.
The exception is cleanup work — removing crowd noise from a fancam or wind from an outdoor take. There, the difference is obvious by design: the audience, the rumble, or the hiss should simply be gone while the music stays intact. If you can't tell immediately, that's a different kind of problem.