Reply to thread

Message: [QUOTE="Queen, post: 90458, member: 27"] Mid-side compression ignores left and right entirely and works instead on a centre channel and a difference channel. The mid is left plus right. The side is left minus right, so it holds only what the two channels refuse to share. Decoding reverses the sum, and left comes back as mid plus side while right comes back as mid minus side. Run both conversions with plain arithmetic, and the output arrives six decibels louder than the signal you started with. Three decibels of trim at each stage keeps that honest. Alan Blumlein worked the matrix out decades before anyone had a plugin folder. Abbey Road wired it into the transformers of its REDD and TG desks, and mastering rooms have leaned on the same sum and difference idea ever since. [HEADING=2]Stereo linking already decides how your image behaves[/HEADING] Every stereo compressor chooses the sidechain before you touch a knob. Linked units read both channels together, so attenuation lands evenly, and the positions of everything inside the mix stay put. Unlinked units let each side act alone. That difference matters more than most people expect. A loud sound sitting off centre can pull one channel down without touching the other, and the image drifts towards the quieter side while the gain reduction lasts. Chris Lord-Alge has argued for years that linked compressors sound more mono, which is a preference rather than a rule. Mid-side operation is a third answer to the same question. You are still splitting the sidechain, but along the centre and the edges rather than the speakers. [HEADING=2]Squeezing the side channel widens and buries at once[/HEADING] Slow settings do the least damage here. A hundred milliseconds of attack with a release somewhere between one hundred and two hundred works on bass-heavy material, and a hundred and fifty milliseconds of release paired with a low ratio suits gentler work. Anything faster starts grabbing the reverb tails that give the sides their depth. Aim for [B]a couple of decibels of transparent gain reduction on the sides[/B], then use makeup gain to lift what is left. Low-level ambience, room tone, and stereo delays all rise with it. The mix sounds bigger without a single width control being touched. There is a bonus buried in that routing. Even with the threshold parked well out of the way, independent gain on each channel hands you a straight balance between the centre and the edges. Plenty of thin masters get rescued with nothing more than that. Watch the bottom end while you do it. Low-frequency content parked in the side channel eats gain reduction and can distort under a short release, which is why plenty of engineers pull the lows out of the sides with a matching equaliser first. [HEADING=2]A mono check exposes an over-boosted side channel[/HEADING] Fold a mix to mono, and the side channel disappears completely. The polarity relationship that carries the width is exactly what cancels when left and right sum together, so every decibel you added to the sides vanishes at the same moment. What survives is the mid, and only the mid. That is the trap. Boost the sides hard enough, and the lead vocal, kick, and bass sitting dead centre get pushed behind everything around them, which reads as impressive on headphones and thin on a phone speaker. Minimum-phase equalisers make it messier, because they shift stereo positions unevenly across the frequency range. Any processor worth using has to give you real control over both streams rather than a single width knob, and [URL='https://goldmidi.com/community/threads/m-media-audio-introduced-the-the-new-yorker-vari-mu-compressor.75872/'][B]a tube-flavoured compressor with separate mid and side controls[/B][/URL] can usually be tested through a demo before money changes hands. Check the fold to mono while you work rather than once at the end, because the damage only shows up after the sides have already been pushed. [/QUOTE]

Name