You are here

Should the side channel always be lower than the mid channel in an analyzer?

For everything after the recording stage: hardware/software and how you use it.

Moderator: Moderators

Should the side channel always be lower than the mid channel in an analyzer?

Postby armans » Tue Dec 01, 2020 10:35 am

I have a mix going at the moment where the song starts out with two guitar parts panned almost hard left and hard right. They do completely different things.

Here is a link to the clip
https://gofile.io/d/Ugwg97

when I open up a spectrum analyzer and compare the mid and side channels overlaying each other I notice that at times the mid channel is softer than the side channel at certain frequencies. Is this a problem?

Finally, here is a reference mix that has two panned guitar parts also very hard panned opposite each other, yet when I open up this file and compare the mid and side channels, the side channel seems neatly tucked below the mid channel at all times and I can't figure out why since there are no central panned instruments at all.
https://gofile.io/d/4EbYDn

thank you!!
User avatar
armans
Regular
Posts: 88
Joined: Sat Jan 11, 2020 2:36 pm

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby Hugh Robjohns » Tue Dec 01, 2020 10:48 am

armans wrote:I notice that at times the mid channel is softer than the side channel at certain frequencies. Is this a problem?

No. As long as it's occasional and not across the whole spectrum it's not a problem. I've monitored your track and I don't see any problem with it.

As you have a vectorscope, you need the 'ball of string' to be circular or narrower horizontally -- so taller than it is wide. If it narrows vertically (so it's wider than it is tall) you have a problem.

Most vectorscopes include a phase meter, and you want the moving pointer to be in the top half between 0 and 1. Occasional brief dips just below zero (towards -1) are tolerable provided they are brief, occasional and only just below the centre. Anything more than that implies substantial phase errors between channels.

...the side channel seems neatly tucked below the mid channel at all times and I can't figure out why since there are no central panned instruments at all.

You're obsessed with MS while ignoring all the other information available to you! :think: MS is only HALF of the full story. You also need to evaluate the LR signals.

Had you done so with your reference track you would have noticed that the guitar is not actually full panned right. :shocked: You can also see that clearly on a vectorscope since the diagonal 'line' representing the guitar is at an angle of less than 45 degrees.

If you mute the right hand channel you can still hear guitar in the left channel, along with the steel drum -- and you obviously wouldn't hear it there if it was fully panned.

The Side amplitude equals the Mid amplitude ONLY for a fully panned signal. In your reference example it's not fully panned, and so the Side remains a reasonably consistent 3dB or so lower than Mid.
User avatar
Hugh Robjohns
Moderator
Posts: 29508
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby blinddrew » Tue Dec 01, 2020 11:08 am

It's fairly common to high pass the sides signal fairly aggressively as well, which will show up strongly on the levels.
User avatar
blinddrew
Jedi Poster
Posts: 12868
Joined: Sat Jul 04, 2015 11:00 pm
Location: York
Ignore the post count, I have no idea what I'm doing...

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby armans » Tue Dec 01, 2020 3:57 pm

Hugh Robjohns wrote:As you have a vectorscope, you need the 'ball of string' to be circular or narrower horizontally -- so taller than it is wide. If it narrows vertically (so it's wider than it is tall) you have a problem.

Ok fantastic, I will look more at the vectorscope. I do like your analogy because it does indeed look like a ball of string :)

Hugh Robjohns wrote:Most vectorscopes include a phase meter, and you want the moving pointer to be in the top half between 0 and 1. Occasional brief dips just below zero (towards -1) are tolerable provided they are brief, occasional and only just below the centre. Anything more than that implies substantial phase errors between channels.

Thanks for the tip!

Hugh Robjohns wrote:You're obsessed with MS while ignoring all the other information available to you! :think: MS is only HALF of the full story. You also need to evaluate the LR signals.

Had you done so with your reference track you would have noticed that the guitar is not actually full panned right. :shocked: You can also see that clearly on a vectorscope since the diagonal 'line' representing the guitar is at an angle of less than 45 degrees.

haha, no I am not obsessed it is just that R & L channels are easier to understand so I don't come asking about it that much. I promise I am not disregarding it at all and I also promise that I did notice that the guitar was not panned all the way to the one side but neither is mine :)

Hugh Robjohns wrote:If you mute the right hand channel you can still hear guitar in the left channel, along with the steel drum -- and you obviously wouldn't hear it there if it was fully panned.

Yes yes, I am glad how thorough you are. Seriously, on other forums sometimes the guys reply before they even read what you have written and you have actually noticed my "mistakes" which makes me smile. I do need to compliment you and this forum because the guys here are really helpful :)

Also, I have no idea what a steel drum is so thanks for that. I must look it up. I thought that that instrument was actually a lap steel guitar playing harmonics of some kind :)


Hugh Robjohns wrote:The Side amplitude equals the Mid amplitude ONLY for a fully panned signal. In your reference example it's not fully panned, and so the Side remains a reasonably consistent 3dB or so lower than Mid.

ooh thanks, another good tip :)
User avatar
armans
Regular
Posts: 88
Joined: Sat Jan 11, 2020 2:36 pm

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby Kwackman » Tue Dec 01, 2020 4:07 pm

They're almost obsolete now, but you might enjoy looking at old fashioned stereo twin PPMs.
They were used in broadcasting, but Zplane do a pretty good software emulator.

https://products.zplane.de/ppmulator

You have two sets of meters, both with two needles.
One meter will have red and green needles, the red is left, and green is right.
The other meter will have white and yellow needles, white for M and yellow for S.

It's great for seeing what happens to each signal when you mess about with the panning.
User avatar
Kwackman
Frequent Poster (Level2)
Posts: 2062
Joined: Thu Nov 07, 2002 12:00 am
Location: Belfast
Cubase, guitars.

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby armans » Tue Dec 01, 2020 4:29 pm

Hugh Robjohns wrote:
The Side amplitude equals the Mid amplitude ONLY for a fully panned signal.


Also, one more thing regarding what you said here. I have voxengo set up on my master stereo bus and I tried soloing and fully panning some mono tracks in my project lto see if the amplitude would match when it was fully panned and it doesn't seem to be happening in this case. Do you have any idea why?

here is a screenshot of voxengo where you can see the side (pink) channel having much more amplitude than the mid (green) channel

https://gofile.io/d/ocsppm
User avatar
armans
Regular
Posts: 88
Joined: Sat Jan 11, 2020 2:36 pm

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby MOF » Tue Dec 01, 2020 4:56 pm

Also, I have no idea what a steel drum is so thanks for that. I must look it up. I thought that that instrument was actually a lap steel guitar playing harmonics of some kind :)

The second link definitely sounds like a steel drum to me too, I suppose it could be a heavily low-pass filtered lap steel guitar but it sounds too percussive to my ears.
MOF
Frequent Poster
Posts: 1402
Joined: Thu Mar 06, 2003 12:00 am
Location: United Kingdom

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby armans » Tue Dec 01, 2020 6:11 pm

MOF wrote:
Also, I have no idea what a steel drum is so thanks for that. I must look it up. I thought that that instrument was actually a lap steel guitar playing harmonics of some kind :)

The second link definitely sounds like a steel drum to me too, I suppose it could be a heavily low-pass filtered lap steel guitar but it sounds too percussive to my ears.

the only reason why I assumed it was a lap steel guitar is because one comes in in the bridge so I thought, makes sense it is just the same instrument in the intro
User avatar
armans
Regular
Posts: 88
Joined: Sat Jan 11, 2020 2:36 pm

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby Hugh Robjohns » Tue Dec 01, 2020 7:27 pm

Kwackman wrote:They're almost obsolete now, but you might enjoy looking at old fashioned stereo twin PPMs.

:-D I still have a mechanical twin-twin PPM and use it alongside the vectorscope. It's a very informative thing once it's understood! And the Zplane software version is remarkably good

You have two sets of meters, both with two needles.
One meter will have red and green needles, the red is left, and green is right.
The other meter will have white and yellow needles, white for M and yellow for S.

Your age is showing! ;) The yellow Sides needle was changed to black with an orange tip in the early 90s. The reasoning being that as the meters aged and the heat from the festoon lamp affected the paint, the white needle yellowed, and the yellow needle bleached, leading to confusion as the two ended up looking much the same (few users noticed that the front needle is always the Mid one (or the left channel)!
User avatar
Hugh Robjohns
Moderator
Posts: 29508
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby Kwackman » Tue Dec 01, 2020 7:32 pm

Hugh Robjohns wrote:Your age is showing!
That's ageist!
(but sadly true....) :shh:
User avatar
Kwackman
Frequent Poster (Level2)
Posts: 2062
Joined: Thu Nov 07, 2002 12:00 am
Location: Belfast
Cubase, guitars.

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby armans » Wed Dec 02, 2020 8:36 pm

armans wrote:
Hugh Robjohns wrote:
The Side amplitude equals the Mid amplitude ONLY for a fully panned signal.


Also, one more thing regarding what you said here. I have voxengo set up on my master stereo bus and I tried soloing and fully panning some mono tracks in my project lto see if the amplitude would match when it was fully panned and it doesn't seem to be happening in this case. Do you have any idea why?

here is a screenshot of voxengo where you can see the side (pink) channel having much more amplitude than the mid (green) channel

https://gofile.io/d/ocsppm

Did you see the image? With it all the way panned the mid and side channel dont look the same at all. Any idea why?
User avatar
armans
Regular
Posts: 88
Joined: Sat Jan 11, 2020 2:36 pm

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby Hugh Robjohns » Wed Dec 02, 2020 10:09 pm

I did, but I don't use Span so I'm not sure what it's displays are doing.

It would probably be less confusing to stick to the vectorscope and/ or straight LR/MS metering, rather than getting bogged down in frequency-related MS balances.
User avatar
Hugh Robjohns
Moderator
Posts: 29508
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Should the side channel always be lower than the mid channel in an analyzer?

Postby armans » Thu Dec 03, 2020 9:20 am

Hugh Robjohns wrote:I did, but I don't use Span so I'm not sure what it's displays are doing.

It would probably be less confusing to stick to the vectorscope and/ or straight LR/MS metering, rather than getting bogged down in frequency-related MS balances.

Ok thanks for getting back to me and for all your help.
User avatar
armans
Regular
Posts: 88
Joined: Sat Jan 11, 2020 2:36 pm