A lot of engineers don’t pay enough attention to timing problems in the mixes.
You can EQ the hell out of the mix but if the groove is wrong, mix will never breathe.
Use a guitar amp emulation on a vocal.
Use a drum compressor on a bass.
Use a vocal reverb on a snare.
Breaking the "intended use" of plugins creates unique textures.
The best sound design happens when you stop following the manual.
Stereo image that works.
Keep sub frequencies (below 120Hz) mono. Allow mid frequencies (120Hz-2kHz) slight width. Open up high frequencies (above 2kHz) for stereo effects. This mimics how natural sound propagates and ensures mono compatibility while maximizing perceived width.
🎚️ SZA lead vocal chain.
Song: Kill Bill
Engineer: Rob Bisel
“There’s a block of five lead vocal tracks, and track 47 (main lead vocal) is like a filtered sample vocal that comes in during the second pre‑chorus. All lead vocals go to the Lead Vocal bus, which is where everything happens in terms of treatments, with a chain of the FabFilter Pro‑DS, Pro‑Q 3, Waves CLA‑3A, Pro‑C 2 and sends to five effect aux tracks, with reverb from Valhalla VintageVerb and Guitar Rig, delay from Soundtoys EchoBoy, and the Soundtoys Little MicroShift and [Waves] MetaFlanger. The reverbs are pretty prominent, but the rest of the auxes are all subtle brush strokes.
These treatments are pretty standard when I work with SZA. I try to keep it in this zone, so when I do a bounce, it doesn’t sound too different song by song."
Source: SoundOnSound Interview.
@natemixing
Instruments occupy many frequencies, you can’t split all instruments between the frequency ranges.
But what you can do is to give one zone to each instrument where it dominates, and cut the conflicting instruments from this exact range (if they try compete for this exact range).
This is one of the key EQ purposes.