Do you use mid/side processing?

[8-bitheroes]

So its something I have known about for a long time but never knew enough to actually utilise it in my mixing or mastering. I recently started using pro q which with its mid/side option makes it really easy to mess around and see what's what. Thing is although I understand what it is and does, its difficult to know when its needed and when it isn't. So if any you guys use this technique share with me your processes and experiences.

[Buoy]

If you make a bus with all the crispy-sounding instruments that have nice high-frequency content (typically everything except bass and kick/snare), put a high-pass filter on it and isolate the side channel from that, then you have an audio equivalent of powdered sugar that you can add reverb/chorus to (or other "pretty" sounds effects) and it will sound delicious

[Oakwood]

I'm stealing this thread: what is mid/side processing?

[Eagleon]

At 7/27/15 07:55 PM, LiquidOoze wrote: I'm stealing this thread: what is mid/side processing?

The mid channel is everything that appears in both left and right channels. The side channel is what appears only in either left or right but not both. The Fruity Stereo Shaper help file explains a lot of what mid/side processing is about, and you can lift the numbers in the last section to accomplish it in various other stereo manipulation dealiewhatsits.

(I've never actually used this, there's so much else to mess with :F)

[Noisysundae]

Every sundae uses it. :P
Much like Buoy said. For my case, I use stereo separation only a bit and focus more on phase delay. It's amazing that just a milisecond delay between left and right can bring you a wider soundstage, isn't it?

I normally use a 1-2 ms phase delay on the drums and bass (except crashes) and > 4 ms on the other instruments. (These can be changed though. Do what you think it's fine.) Anyway, don't put too many instruments in the same delay amount and side or your music won't sound wide enough. Phase inversion on one side is an option too especially when your bass and kick are having a fight. (beside compression on low frequencies)

Remember, if you're going to use these effects in mono e.g. mixing music for some objects in a game, they'll make phase cancellations which will ruin your entire piece.

At 7/27/15 09:27 PM, Eagleon wrote:
...The Fruity Stereo Shaper help file...

Your link is dead. Use this instead.

[MetalRenard]

Do I?
Always. Mid/side mixing is key to having that huge sound.
I will share info later. Sleepy now. Night night. u_u

[MetalRenard]

Double post time - Morning everyone. Edit: Thanks for saving me Lich!
Some of what I say might not apply to you because I mix rock / metal / classical and you seem to stick to electronic genres of music. Having said that, concepts definitely apply so I hope you can take something away from it.

Rock/metal
You have to start by making sure everything sounds good in mono before you go to mid/side because mid/side creates the illusion of space using the stereo spectrum but if the consumer listens on a phone for example, with a shitty mono speaker (I HATE THOSE GUYS and they're multiplying like a disease in the audio world) then your mix still needs to work for them.
I consider mid/side mixing among the final few steps for this reason.

You have to treat each frequency band and instrument differently. For example, I would never make a bass guitar wide using this, however I may slightly widen the low frequencies (below 130hz) of my guitar to give the bass more space and to keep that area cleaner. Consider that my guitar has a high pass filter at about 100hz anyway.
I tend to reduce the mid content for electric guitars too, to give space to other instruments (piano etc).

Now for the high frequencies, I usually use mid/side to make my cymbals sound wide, and I also widen the high guitar frequencies quite a lot. This gives space for vocals. Don't use phase to widen stuff, this will only sound good in a few situations and usually sounds terrible on headphones (because your brain using this delay of 1ms or less to detect WHERE something is coming from. If it comes in early it's REALLY confusing!). It creates all kinds of issues elsewhere too (tried listening on a TV?! Those guys will spit all over your "perfect" mix).

Anyone looking for a FREE tool to do this with, try Melda Production's free bundle. Their EQ can be set to mix mid OR side. Just put two on each track to mix each separately! :D However I use Ozone 5.

[8-bitheroes]

At 7/28/15 04:15 AM, MetalRenard wrote: Double post time - Morning everyone. Edit: Thanks for saving me Lich!
Some of what I say might not apply to you because I mix rock / metal / classical and you seem to stick to electronic genres of music. Having said that, concepts definitely apply so I hope you can take something away from it.

Thanks for going into a bit of depth, deffinitly helpful info there

[8-bitheroes]

At 7/27/15 07:55 PM, LiquidOoze wrote: I'm stealing this thread: what is mid/side processing?

Give it back

[Oakwood]

At 7/28/15 05:54 AM, 8-bitheroes wrote:

At 7/27/15 07:55 PM, LiquidOoze wrote: I'm stealing this thread: what is mid/side processing?

Give it back

Sorry man, here ya go.

So mid/side processing is pretty much the same as stereo seperation? Yeah I do that often, but not as much as I actually shouldI think.

[thebitterroost]

I never have. Gotten by pretty well so far just with ordinary stereo panning techniques.

[DL2Electron]

@8-bitheroes

I just discovered M/S processing aswell just a few months ago.

People seem to mainly use it during the mastering stage because it gives surgical control to tweak the sound. Like you can cut the low ends in the side channel to get rid of some mud. Or you can boost the frequency range where the vocals are in the mid channel if you want to keep the stereo/side channel untouched. M/S processing just doesn't end there though. There's M/S compression, M/S multiband compression, M/S FX etc etc... Yeah, you can go pretty crazy with M/S processing like @MetalRenard said. On FL Studio you can setup a couple of stereo shapers to make a M/S chain inside of patcher. Takes a lot of CPU though.

[Eagleon]

At 7/29/15 04:46 PM, DL2Electron wrote: @8-bitheroes

I just discovered M/S processing aswell just a few months ago.

People seem to mainly use it during the mastering stage because it gives surgical control to tweak the sound. Like you can cut the low ends in the side channel to get rid of some mud. Or you can boost the frequency range where the vocals are in the mid channel if you want to keep the stereo/side channel untouched. M/S processing just doesn't end there though. There's M/S compression, M/S multiband compression, M/S FX etc etc... Yeah, you can go pretty crazy with M/S processing like @MetalRenard said. On FL Studio you can setup a couple of stereo shapers to make a M/S chain inside of patcher. Takes a lot of CPU though.

Patcher is single threaded and tends to choke on more complex tasks, which is really silly because complex tasks are what it's supposed to be for :( You may or may not be better off routing a mixer channel into two extra, putting a stereo shaper on each to get mid/side, and routing them both into yet another to decode.