Matching Reference Loudness
Bring the reference down to the same loudness first, or the ear is only comparing volume.
Related
Bring the reference down first
Play the reference against your own version and nine times out of ten the reference wins, because it is simply louder. Three dB of level is enough to make the low end seem fuller, the vocal closer and the detail richer. Match the two before comparing tone; skip that step and every later judgement is unreliable.
Match integrated loudness
Match on integrated loudness over the whole track, within 0.1 dB. If the reference sits at -9 LUFS, bring it down to -14 before listening; if your version sits at -16, bring it up to -14. Use short-term loudness to compare sections, especially how full the chorus really is. Watch true peak separately and do not break -1 dBTP just to make the loudness match. The match changes monitoring level only: not the file, and not a limiter on the bus chasing the reference.
A comparison table once levels agree
| Dimension | What to hear | What not to conclude |
|---|---|---|
| Low end | Which is fuller below 100 Hz, and whether it muddies | Fuller is not better |
| Midrange | How vocal and lead sit front to back | Different arrangements place them differently |
| High end | Air above 8 kHz | Bright is not clear |
| Transients | Drum attacks and bass onsets | A remaining gap after matching is real |
| Width | Difference between mono and stereo | A style choice, not a right answer |
| Dynamics | The drop between verse and chorus | The reference may already be squashed |
Compare one dimension at a time, ten seconds per excerpt, returning to the same phrase. When the reference still seems better, write down whether it is low-end weight, vocal position or transients, then go back to the mix rather than pushing the bus louder.
What level matching cannot fix
The reference differs in arrangement, performance and recording space, so matching loudness only removes the volume bias; it cannot turn different forces into one ruler. A heavily compressed reference cannot set your dynamic target, and a live recording cannot dictate your sense of space. Another common error is reading the reference master backwards to set mix levels, which reverses the order of work and only makes the result tighter.
Listen in the library
All tracks from midicn-lib — inline badge is the license tier (C1 commercial / C2 non-commercial / C3 study only).
Along the same line
Sources
- ITU-R BS.1770 与 EBU R128 关于整合响度、真峰值与平台归一化的定义。
- 电平对齐后的参考比较方法，依据听感训练与混音实践。