You are here

Frequencies in the stereofield of the mix

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

Moderator: Moderators

Frequencies in the stereofield of the mix

Postby cremedonut » Mon Apr 05, 2021 4:23 pm

Hey there,

I want to take a theoretical approche to assigning frequencies to specific points in the Stereo field from an abstract persepctive.

To make a mix sound wide there are different factors to be accounted for. One of them being paning of different signals. It can be advantages to mono all frequencies from 80-120hz as fas as I understand it. Frequencies between 800hz and 2.5khz are easily locateable. So theoretically speaking frequencies above 2.5khz are to be panned hard right and left. I am aiming for an abstract reference, assigning frequencies to a place between center and right/left. Is there some kind of mathematical way to calculate this?

Whether or not this kind of perspektive is useful in practical applications is debatable. For experimentation I, however, would like to have an abstract reference.

Please share your thoughts!
cremedonut
Poster
Posts: 13
Joined: Mon Apr 05, 2021 4:05 pm

Re: Frequencies in the stereofield of the mix

Postby controlcentral » Mon Apr 05, 2021 6:18 pm

Although you are seeking a mathematical, abstract answer, I will offer instead a practical, empirical suggestion:
Assuming you are "in the box", and to keep the list of possibilities brief, get a trial of Nugen Audio's Stereo Placer and see what works for you IRL.
controlcentral
Posts: 2
Joined: Mon Apr 05, 2021 6:00 pm

Re: Frequencies in the stereofield of the mix

Postby cremedonut » Mon Apr 05, 2021 11:23 pm

That is a really nice plugin. I will make use of their offer to try the trail version, as soon as I get the time to focus on it. Thank you for this tip! Eventually it comes down to experience and what works for me in a particular situation, I agree. Painting by numbers only goes so far. That is the sound position on this question.

Still, as a reference to contrast what I'd do intuitively it would be really nice to have a blueprint. Especially because I am interested in experimenting on things that are rather unorthodox and aren't possible to edit/move around on a whim.

When aiming for a wide sounding mix, a mono lowend and a increasingly spread highend is desirable. At a certain point a further spreading of the highend is contraproductive, since the human ear/brain cannot locate the information.

A mono-maker may be applied from 80 to 100hz to make that range completley centric. From that point on the information should be relativly in the middle, still. With the information between 800hz and 2.5khz being easily locatable this is the critical range for spreading information in the stereofield. Apart of this being a fundamentaly flawed approach, since it bypasses all IRL feedback and practice, is this plausible so far, or am I getting something wrong?

The relation, then, between hertz and sidepositon value is not linear I assume. This relation is what I'm interested in most. I'd like to have a rough calculation on this relation if possible. This is for most practiable applications completely irrelevant, however if working on specific experimental approaches that have a complex setup this would be handy for me.
cremedonut
Poster
Posts: 13
Joined: Mon Apr 05, 2021 4:05 pm

Re: Frequencies in the stereofield of the mix

Postby Kwackman » Tue Apr 06, 2021 8:27 am

cremedonut wrote:When aiming for a wide sounding mix, a mono lowend and a increasingly spread highend is desirable.

You might find it desirable... ;)
User avatar
Kwackman
Frequent Poster (Level2)
Posts: 2196
Joined: Thu Nov 07, 2002 1:00 am
Location: Belfast
Cubase, guitars.

Re: Frequencies in the stereofield of the mix

Postby Bob Bickerton » Tue Apr 06, 2021 8:39 am

Given your suggested approach Is unorthodox and experimental and not something dwellers of the orthodox world are likely to have explored, or are interested in doing, it strikes me that you do indeed need to experiment and not calculate.

Easy to do if you’re working in the box.

Bob
User avatar
Bob Bickerton
Jedi Poster
Posts: 4189
Joined: Fri Dec 20, 2002 1:00 am
Location: Nelson, New Zealand
http://www.bickerton.co.nz

Re: Frequencies in the stereofield of the mix

Postby Hugh Robjohns » Tue Apr 06, 2021 12:29 pm

cremedonut wrote:I want to take a theoretical approche to assigning frequencies to specific points in the Stereo field from an abstract persepctive.

What's the theory behind this approach?

It can be advantages to mono all frequencies from 80-120hz as fas as I understand it.

Yes if you're cutting a vinyl record. No if you want a spacious sound character.

So theoretically speaking frequencies above 2.5khz are to be panned hard right and left.

Why? What's the theory?

I am aiming for an abstract reference, assigning frequencies to a place between center and right/left. Is there some kind of mathematical way to calculate this?

No. Because there is no sense to it.

You could achieve the same effect very simply by setting up the 'fake stereo' treatment of taking a mono signal and duplicating it, feeding one part into the Mid input of a mid-side decoder, and high-pass filtering and delaying the other part before feeding it into the Side input of the decoder.

The result is a pair of orthogonal comb-filters in the left-right channels, where different frequencies have different pan amounts in a pseudo-random arrangement (dependent on the delay value). This creates an illusion of stereo width but without any recognisable imaging as such. It's abstract...

Whether or not this kind of perspektive is useful in practical applications is debatable.

Wouldn't take much debate. Random scattering of different frequencies around the sound stage will create a blurred, unfocused, spread of sound. There are a few applications where that might be appropriate or acceptable, and there are ways of achieving it quickly and easily (As outlined above).
User avatar
Hugh Robjohns
Moderator
Posts: 30685
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound
In my world, things get less strange when I read the manual... 

A bespoke can of worms [Re: Frequencies in the stereofield of the mix]

Postby controlcentral » Tue Apr 06, 2021 3:01 pm

cremedonut wrote:
...as a reference to contrast what I'd do intuitively it would be really nice to have a blueprint. Especially because I am interested in experimenting on things that are rather unorthodox and aren't possible to edit/move around on a whim.

Not being able to move things around 'on a whim' suggests some sort of fixed, physical sound installation, but you also imply that your pedagogical inquiry involves a stereo mix, i.e., loudspeakers, which is a bit confusing. So unless you are able to articulate specifically what you are after in a more precise manner, the replies you receive may be of little help and will probably serve to annoy the community.
From what little I understand of what you are actually asking, I'm sorry to be the bearer of bad news: no blueprint is to be had. The equation involves much more than left, right, and frequency; there is an inter-aural intensity (L-R volume difference), as well as inter-aural time differences ( when the sound/ frequency reaches each ear), phase differences, and a HRTF (head-related transfer function).
Furthermore, these phenomenon may be perceived differently/ inaccurately depending on the playback method (i.e., binaural recordings on stereo speakers), then moving on to the Ambisonics WXYZ format and its decoding implementation (for example Harpex), and even further, systems like Occulus VR and Dolby Atmos... :crazy:
Not to be a shill for Nugen, but perhaps it would be helpful to read the manual of another one of their sound-field related plugins (which handily addresses many of the above issues in a tidy package) in order for you to get a handle on some of the issues involved here.

All that said, this forum is geared ( :lol: ) toward working audio professionals-- it's possible you may wish to peruse more theoretical journals and papers in your quest, a totally random one of which I have linked for you here:
An Introduction to Audio Spatialization
Best of luck in your further pursuits!
controlcentral
Posts: 2
Joined: Mon Apr 05, 2021 6:00 pm

Re: Frequencies in the stereofield of the mix

Postby RichardT » Tue Apr 06, 2021 7:22 pm

cremedonut wrote:Hey there,

I want to take a theoretical approche to assigning frequencies to specific points in the Stereo field from an abstract persepctive.

To make a mix sound wide there are different factors to be accounted for. One of them being paning of different signals. It can be advantages to mono all frequencies from 80-120hz as fas as I understand it. Frequencies between 800hz and 2.5khz are easily locateable. So theoretically speaking frequencies above 2.5khz are to be panned hard right and left. I am aiming for an abstract reference, assigning frequencies to a place between center and right/left. Is there some kind of mathematical way to calculate this?

Whether or not this kind of perspektive is useful in practical applications is debatable. For experimentation I, however, would like to have an abstract reference.

Please share your thoughts!

I don’t know of any reference for you. If you want to try it I would calculate the pan position using the log of the frequency, normalised so that 20Hz is centre (pan 0) and, say, 10kHz is extreme left and right (pan 100, or whatever the max is your DAW).

In fact, you’ll be working with frequency bands I expect, so use the mid frequency of the band.

The results will be very strange!
RichardT
Frequent Poster
Posts: 751
Joined: Fri Aug 13, 2004 12:00 am
Location: London UK

Re: Frequencies in the stereofield of the mix

Postby cremedonut » Tue Apr 06, 2021 8:42 pm

Hugh Robjohns wrote:
cremedonut wrote:I want to take a theoretical approche to assigning frequencies to specific points in the Stereo field from an abstract persepctive.

What's the theory behind this approach?
Human perception locating frequencies in a nonlinear nature, roughly speaking.

Hugh Robjohns wrote:
It can be advantages to mono all frequencies from 80-120hz as fas as I understand it.
Yes if you're cutting a vinyl record. No if you want a spacious sound character.
True, I'm, however, overdubbing and not going for a realistic space but artificial wideness. With this in mind, my idea is to have even amounts of low end in both speakers for more punch, since the lowend is hardly locateable.


Hugh Robjohns wrote:
So theoretically speaking frequencies above 2.5khz are to be panned hard right and left.

Why? What's the theory?
In theory the highend is only easily locateable for a human being up to a certain point.
cremedonut
Poster
Posts: 13
Joined: Mon Apr 05, 2021 4:05 pm

Re: A bespoke can of worms [Re: Frequencies in the stereofield of the mix]

Postby cremedonut » Tue Apr 06, 2021 9:07 pm

controlcentral wrote:(...) Unless you are able to articulate specifically what you are after in a more precise manner, the replies you receive may be of little help and will probably serve to annoy the community.
I see your point. Here is a practical example:

I have a convetionally mixed rock-drum with a centered mono bass-guitar. On top of that I have two different guitars playing the same rhythm. Both guitars are split into multiple channels running through different gear, causing different types of coloration. Each coloration is then filtered in different ways (different IRs for example) adding even more shades of coloration. All those many shades of coloration are then bandpassed and put next to eachother.

A simplfied example:
Lowpass 150hz mesa Rectifier
Bandpass 150hz-6khz Plexi
Highpass 6khz headphoneamp

All those bandpassed colorations I want to spread throughout the stereofield.
cremedonut
Poster
Posts: 13
Joined: Mon Apr 05, 2021 4:05 pm

Re: Frequencies in the stereofield of the mix

Postby cremedonut » Tue Apr 06, 2021 9:13 pm

RichardT wrote:(...) If you want to try it I would calculate the pan position using the log of the frequency, normalised so that 20Hz is centre (pan 0) and, say, 10kHz is extreme left and right (pan 100, or whatever the max is your DAW).

In fact, you’ll be working with frequency bands I expect, so use the mid frequency of the band.

The results will be very strange!

I have added this to my notes and will give it a try. Thank you very much. And thanks alot for all the kind answers in this thread!
cremedonut
Poster
Posts: 13
Joined: Mon Apr 05, 2021 4:05 pm

Re: Frequencies in the stereofield of the mix

Postby Mixedup » Wed Apr 07, 2021 8:34 am

Only tangentially related, but it occurred to me that you might also like to play with Melda's MSpectralPan plug-in.
User avatar
Mixedup
Jedi Poster
Posts: 4455
Joined: Wed Sep 03, 2003 12:00 am
Location: Cambridgeshire, UK

Re: Frequencies in the stereofield of the mix

Postby cremedonut » Wed Apr 07, 2021 9:44 am

Mixedup wrote:Only tangentially related, but it occurred to me that you might also like to play with Melda's MSpectralPan plug-in.

Yes, this looks promising. Thanks!
cremedonut
Poster
Posts: 13
Joined: Mon Apr 05, 2021 4:05 pm

Re: Frequencies in the stereofield of the mix

Postby CS70 » Wed Apr 07, 2021 11:49 am

cremedonut wrote:True, I'm, however, overdubbing and not going for a realistic space but artificial wideness. With this in mind, my idea is to have even amounts of low end in both speakers for more punch, since the lowend is hardly locateable.

Just in passing: you normally have even amounts of low end in both speakers because the low end is usually set to mono (which is defined as even amounts of something in both speakers - aka the two channels of a stereo recording).

The only way not to have an even amount of low end in both speakers is to explicitly pan, say, a bass guitar on one side. But that is rather pointless, and it doesn't mean that you would hear the bass coming from one side, at least in a regular sized room.

That's because the fact that the low end is not directional is not a property of the panning, but a property of the wave length of low frequencies with respect to an average room size and how the reflections behave. A "wave" is really a movement of expanding air with a spherical boundary, centered on the speaker. By the time a *full* wave has been produced, its "beginning" is already bouncing around on the walls and reaching you from all directions - that's why it's not directional in a room.

Panning the low end will simply force one speaker to work more with respect to the other (and therefore limit the maximum total amount of energy that can be pumped in the room), but the bass would still be non-directional. In extreme circumstances, say a stadium, the panning would simply change the timing of the low frequency depending on where you are listening (because in a stadium the low end frequency waves are much smaller than the reflective walls, so they become more directional).

On say headphones, the panning would be immediately perceivable - since the sound is literally injected in your ears and little reflections are involved.
User avatar
CS70
Jedi Poster
Posts: 7746
Joined: Mon Nov 26, 2012 1:00 am
Location: Oslo, Norway
Silver Spoon - Check out our latest video and the FB page

Re: Frequencies in the stereofield of the mix

Postby cremedonut » Wed Apr 07, 2021 8:43 pm

@CS70 Yes, that is my understanding as well. Mick Guzauski uses a mono-maker on every mix, as far as I know. Might not be the right choice in every situation, but seem to work for the stuff he is doing.
cremedonut
Poster
Posts: 13
Joined: Mon Apr 05, 2021 4:05 pm

Next