Mid-side process problem.

N

nothingness

Member
I've been struggling with the mid-side process. Someone recommended me a Youtube video and I gave it a shot.



The thing is that the mid issue was solved and the stereo field has gone after that. This is a pain in the ass. How could you solve it?
 
Hard to say without knowing how you have processed the mid/side channels. You’d have to make some pretty drastic cuts in the side channel for the stereo field to be affected that much. What kind of curves did you apply?
 
This whole process sounds suspiciously like a variant of the old "Mono reprocessed for stereo" that was prevalent in the 60s. In this case, it's not really M/S which I have found can have a really nice stereo field. Its a more of a pseudo M/S which might have some slight similarity to the real thing. A whole lot of knob twisting to fake what can be easily done with the right equipment.

Then again, if you are doing the recording via direct in, you have no "/S" to work with, so that's probably your only option.
 
You have to be subtle with mid/side processing. I try not to use it unless i am struggling tooth and nail to get a little space.

I am assuming you are only talking about M/S processing in general, and nothing to do with recording with the M/S technique.

If using M/S processing in mastering for enhancement and you are doing big EQ moves, your mix balance is probably off. I have stopped using it almost all together, I don't even use it to mono the low end anymore because even high passing the side channel as low as 80-100hz sounds worse to me. M/S techniques can be very valuable to a mastering Engineer who does not have access to your individual tracks but you do.

It don't know why it sounds worse, I can't put my finger on it. Could just be slight phasing? I doubt in reality I can hear much of a difference by high passing the sides at 80/100hz because I mix with a mono bottom end anyway, so the high pass should essentially be doing nothing. but with a blind A/B test I really just do prefer it in bypass.

I didn't watch the video but a lot of people use M/S EQ on the master to duck a little low mid in mid channell, and a bit of air in the side channel, but the sound always sucks for me, the sides become too much, always. Even when pro's show it in videos as an example. If you use ProQ3 which most do then make those EQ points dynamic for more transparancy.

Some people will use M/S to create space in the rhythm guitar for the lead vocal (reduce mid in classhing frequencies to vocal because vocal is more than likely occupying mid channel), but you can do a better job using the more traditional tools that we have always used. There are so many other more basic options. I think with m/s techniques, if you can, always make it dynamic and sidechain making SUBTLE moves.

I asked in the academy I'm in if M/S processing is necessary for a professional mix. the big answer I got back was a resounding NO
 
Last edited:
What makes you think you actually need M/S processing? I do it fairly regularly, but only because I'm often mixing a trio situation in which the bass needs to be panned. In that case I'll high pass the side at about 300 Hz. But otherwise, I don't generally bother to do any M/S processing.
 
When anyone mentions M/S processing I think predominently they are talking about Mid/Side EQ. The M/S plugins normally just encode the signal so you have individual volume control for the sides and mid channel independently.

I did watch that video posted above, and this was all about M/S EQ. As much as I love that fabfilter guy and respect him, he is just showing off the ProQ3 to achieve most of the same results as what my stock logic plugins can already do, and are designed for. What with the stereo width tools, which is just EQing the left/right exactly opposite.

It's interesting to learn about turning down the mid channel slightly for a clearer mono mix because it will lose more volume when summed to mono, but with the hard panned stuff dropping 3db ish when summed to mono anyway (pan law), I am not sure I would ever use it much in practise, or it may end up just being too quiet. But In electronic music it would probably be way more useful. More ear candy elements to go crazy with.

Personally I don't want to get involved with it until I am at a good enough skill level to get the absolute most I can with standard EQ and width, and then If I want that extra I can use the more advanced processing. Most, if not all of the pro mixes I have seen there is no m/s processing involved. The thought of delving into that kind of specific fine tuning of the stereo field scares me at the moment. I would think of it as more of a mastering thing. But M/S processing is very important for a mastering engineer, it allows a certain element of control of the instruments in a mix. Bob Katz uses it turn turn up some 200hz in the mid channel if the vocal is a little buried, this brings up the vocal in level a bit with minimal side effects to the other surrounding instruments being brought up in volume also. And panned drums or hats, it's easy to then turn down the highs in the left of right channel to essential give a mastering engineer a hi-hat fader.
 
Oh and yeah! your daws standard EQ may allow you to switch between mid or side EQ. It will probably become a standard for all daws in short time.
 
Mr. ROUSH said:
Hard to say without knowing how you have processed the mid/side channels. You’d have to make some pretty drastic cuts in the side channel for the stereo field to be affected that much. What kind of curves did you apply?
Click to expand...
If you watch the video around 7:00, he describes how to apply it. That's what I tried.
 
TalismanRich said:
This whole process sounds suspiciously like a variant of the old "Mono reprocessed for stereo" that was prevalent in the 60s. In this case, it's not really M/S which I have found can have a really nice stereo field. Its a more of a pseudo M/S which might have some slight similarity to the real thing. A whole lot of knob twisting to fake what can be easily done with the right equipment.

Then again, if you are doing the recording via direct in, you have no "/S" to work with, so that's probably your only option.
Click to expand...
Really? I didn't realize that. Btw, I'm using only virtual instruments.
 
JamEZmusic said:
You have to be subtle with mid/side processing. I try not to use it unless i am struggling tooth and nail to get a little space.

I am assuming you are only talking about M/S processing in general, and nothing to do with recording with the M/S technique.

If using M/S processing in mastering for enhancement and you are doing big EQ moves, your mix balance is probably off. I have stopped using it almost all together, I don't even use it to mono the low end anymore because even high passing the side channel as low as 80-100hz sounds worse to me. M/S techniques can be very valuable to a mastering Engineer who does not have access to your individual tracks but you do.

It don't know why it sounds worse, I can't put my finger on it. Could just be slight phasing? I doubt in reality I can hear much of a difference by high passing the sides at 80/100hz because I mix with a mono bottom end anyway, so the high pass should essentially be doing nothing. but with a blind A/B test I really just do prefer it in bypass.

I didn't watch the video but a lot of people use M/S EQ on the master to duck a little low mid in mid channell, and a bit of air in the side channel, but the sound always sucks for me, the sides become too much, always. Even when pro's show it in videos as an example. If you use ProQ3 which most do then make those EQ points dynamic for more transparancy.

Some people will use M/S to create space in the rhythm guitar for the lead vocal (reduce mid in classhing frequencies to vocal because vocal is more than likely occupying mid channel), but you can do a better job using the more traditional tools that we have always used. There are so many other more basic options. I think with m/s techniques, if you can, always make it dynamic and sidechain making SUBTLE moves.

I asked in the academy I'm in if M/S processing is necessary for a professional mix. the big answer I got back was a resounding NO
Click to expand...
Wow, thanks for the details. Actually, I've been struggling with exactly what you mentioned. The rhythm guitar and chorus parts. I moved it slightly and removed the side channel curves as you advised and it worked.
Thank you!
 
bouldersoundguy said:
What makes you think you actually need M/S processing? I do it fairly regularly, but only because I'm often mixing a trio situation in which the bass needs to be panned. In that case I'll high pass the side at about 300 Hz. But otherwise, I don't generally bother to do any M/S processing.
Click to expand...
Aha... most of the time, it is stereo rhythm guitar and chorus parts.
 
nothingness said:
Aha... most of the time, it is stereo rhythm guitar and chorus parts.
Click to expand...
Right, but what is the problem you're trying to solve?

For example, the reason I use it when I have panned bass is that in headphones it can be rather uncomfortable to have low frequencies panned hard to one side. In real life, bass frequencies are almost always heard with both ears, but headphones isolate each ear from the opposite channel. High passing the difference (side) channel at 300 Hz simulates the natural cross bleed of human hearing. A lot of people listen on earbuds and I want their listening experience to be positive. It hardly affects listening in speakers so there isn't much of a downside.

Are your "stereo" parts double tracked or are you trying to artificially make stereo out of mono tracks? Artificial stereo is usually a poor substitute for double tracking.
 
nothingness said:
If you watch the video around 7:00, he describes how to apply it. That's what I tried.
Click to expand...
Gotcha. Well, my best guess is that the processing isn’t needed in your example. If you’re making a few cuts to the side channel, and the stereo field is disappearing, then the element may not have much information on the side channel to begin with. Try using the technique on something else to see if you get the same issue. Try it on something that you know has a lot of information on the side channel. Maybe, you just need to be more subtle with it. Get lots of practice on multiple elements to give yourself a frame of reference.
 
You must log in or register to reply here.
Back
Top