After reading that in ProQ 3 I need to multiply every value by 1.41 (precisely sqrt(2)) to get a compatible setting, I got a bit confused and I've spend half the day comparing macOS EQ solutions.
I made a log sweep .wav file, 24/96/mono.
I took oratory1990's Sennheiser HD650 Optimum settings, and I applied:
- FabFilter ProQ, with 1.41 correction
- FabFilter ProQ, without correction
- EqualizerAPO (via benchmark.exe, 16-bit only)
- AUNBand from Apple (included free in macOS)
Then I compared them all using DeltaW.
Results:
AUNBand ~ EQApo: 77 dB / 71 dBA, gain error max -35 dB
ProQ corrected ~ EQApo: 80 dB / 73 dBA, gain error max -37 dB
ProQ corrected ~ AUNband: 85 dB / 94 dBA, gain error max -37 dB
---
ProQ non-corrected ~ EQApo: 19 dB / 19 dBA, gain error max -3.6 dB (!!)
I'm just starting on EQ-ing, but I guess this tests showed that:
- ProQ needs the 1.41 Q correction!
- EQ-s produce kind of similar results, but not very perfectly.
- The macOS included AUNBand is doing a fine job, at least it is more similar to ProQ then EQApo. It might not necessarily mean good nor bad though.
Is there a reference 10 band parametric EQ which we can use to compare with?
If someone needs it, I'm happy to share my files or settings I used.
I used DeltaW with the following config and I selected Yes when it asked for "simple sine wave" (
@pkane is this right?):
View attachment 158276