The Wind SpedeMix

[Spede]

Hey Spede - how *did* you get so much snap in the snare with this track?
The top snare track is really quite 'muddy'.
I tried EQ (which brought out the hihat), biasing the balance in favour of the bottom mic (which made it rattly), even throwing them out of phase to reduce low end (which just didn't work) - your snare actually sounds like a snare should sound!
Well done!

Thanks If it's any help I can broadly describe what I did. I simply had the goal in my mind when I was tweaking the snare channels.

The snare top is quite distorted. I generally use small amounts of Digidesign Lo-Fi for drum transient distortion since it's character is somewhat related to what Neves and APIs do for drum transients when driven hard (although Lo-Fi has a much bigger danger making the signal little mushy so I have to be really careful and precise with it.). I've added some high end before distortion; Wide bell boosts worth over 10 (closer to 15) dBs for area covering 4-12 kHz. This one needed more than usually. Then a tiny amount of tape saturation to it. The only way I can hear tiny AB is against compressor; a small amount of tape actually make the drum louder when the snare comes from the output of the compressor. After these there was maybe little more added high end before gating the signal quite fast and aggressively; one can barely hear the next hihat hit when the track is soloed. Then just a small amount of compression (only 3 dBs) with a slow attack and a fastest release. I use Waves SSL channelstrips for drums, so that's the drum comp/gate.

The goal is to make the snare top as bright as possible without making the hihat too problematic, and combination of sawing off the transients with dist/tape to that small amount of slow atk./fast rel. compressor brings a certain kind of constrained but punchy low end from the snare. I believe punchyness comes from correct phasing (with everything else) while the crack comes from the exaggerated and distorted high end. I personally believe you can have both

The snare bottom acts in a very different kind of role: sustain. I've expandend it right off the bat, before any distortions with an EQ:ed sidechain so the expander triggers even from the ghosts, but not at all from cymbals. However because of the high bleed it did open from the kick hits somewhat despite the highpass in the sidechain (once again it the SSL channelstrip expanding it. Then there's the same distortion stuff than in the top channel (but little less Lo-Fi and more tape saturation to make it "softer". Finally the SSL channel compressor at the end is chugging good 10 dBs (with a fast attack and fastest release) out of those amplified, distorted "strong hits" while the ghosts become almost as loud as the strong hits. Bleed surfaces approximately to the same level as it was initially.

So in the end I've simply shaped the strong hits to resemble a tight white noise generator-like sound (with minimal attack) and brought up the ghosts to get them audible in the mix overall. I could've just compress the hell out of snare bottom but that would have been little messy. The reason for distorting the transients is to make the compression more even, so it's not all over the place.

On the drum parallel bus there was a Neve 33609 plugin modelling with the fastest release doing 6-8 dBs of gain reduction. It's overall level was pretty much the same as the uncompressed bus, so it could be said it was quite 50/50

The reason why I could get away with such obscene amount of processing on the tracks itself was because of the good drummer. My mentor had a habit to say: "The more consistent the drum performance, the more you can mangle them" The reason for sample usage (at least surely for me and I strongly believe for others too) is that you can do this kind of crazy shaping without having to worry about hihat leakages and hit inconsistencies.

Welp, that was a pretty long-winded answer.