Page 1 of 2
Pan Law
Posted: Sat Jan 13, 2024 6:26 pm
by ColinP
Today I've been looking at pan laws again as Adroit Custom now includes modular mixing elements.
In the graph below blue is constant power, red is the square root approximation, green is linear and magenta is the BBC/SSL -4.5dB compromise.
Apart from wide agreement about linear being a poor choice what are people's opinions on the merits of the others?
I generally use the square root approximation but as my hearing is asymmetrical I can't really make a good judgement.
Is it worth offering the user options or can I get away with a single law?
- panlaws.png (246.47 KiB) Viewed 45187 times
Re: Pan Law
Posted: Sat Jan 13, 2024 9:33 pm
by utdgrant
No firm preference for pan laws, TBH. I think you could probably settle on one of the popular standards and use it throughout.
However, if you're discussing crossfading two inputs into one output (e.g. wet/dry mix), then I find linear to be the 'correct' solution. The stock CA Crossfade module outputs 100% of both inputs in the centre position. This really annoys me, to the extent that I might be forced to create a free linear version.
- StockCrossfadeTransferFunction.jpg (101.88 KiB) Viewed 45179 times
Re: Pan Law
Posted: Sat Jan 13, 2024 10:15 pm
by UrbanCyborg
I'd say it doesn't usually matter, so long as the method is C2 continuous. For mixing, I'd think the constant-power version would be best, although a close approximation would probably be okay.
Reid
Re: Pan Law
Posted: Sun Jan 14, 2024 2:31 am
by borkman
For panning, I personally don't think it matters too much.
Mixing wet/dry seems more contentious. I did some research for Full Thickness Tear in the plugin world and found that functionality, and opinions of that functionality, varied from plugin to plugin. Linear, sine law (and its cousins), the two knobs in one thing that Grant mentioned, and other approaches were represented. I ended up offering a choice, but I'm not sure that's something I would do in the general case.
Maybe two separate knobs? That has its own complications, but at least it's flexible.
¯\_(ツ)_/¯
Re: Pan Law
Posted: Sun Jan 14, 2024 2:10 pm
by ColinP
Thanks everyone, that's been very useful feedback. I now feel confident that restricting pan law to constant power is an acceptable limitation for improved ergonomics.
I'm trying to simplify things as the menu system in Adroit Custom is approaching labyrinthine.
There's a pre-built mixer strip object that pan law choice wouldn't have had much impact on but one of the generalized mapping options (alongside remote control, CC and CV ) is VCA Pair. This mapping option attempts to offer just about every type of VCA based control functionality via up to 32 VCA pairs in the Custom IO banks.
The following shows the relevant strand in the menu system when a knob mapped to a VCA pair is right-clicked and it's already complicated enough without offering a choice of different panner types...
- VCAPanMenu.png (114.09 KiB) Viewed 44305 times
Re: Pan Law
Posted: Sun Jan 14, 2024 2:21 pm
by ColinP
utdgrant wrote: ↑Sat Jan 13, 2024 9:33 pm
The stock CA Crossfade module outputs 100% of both inputs in the centre position. This really annoys me, to the extent that I might be forced to create a free linear version.
That explains why I got slightly weird behaviour the last time I used the CA Crossfade module.
Go for it Grant!
Re: Pan Law
Posted: Sun Jan 21, 2024 4:26 pm
by ColinP
Many thanks for your free linear crossfader module, Grant!
Just looking at how CA handle pan law and it seems their 6-input Stereo Mixer and Panner modules behave in the same odd way as their Crossfader module.
When in the center each channel is at 100% and if we pan in either direction one channel remains at 100% with just the other channel reducing.
You'd expect both channels to be -3 dB, -4.5 dB or - 6 dB down in the center position if the pan law was one of the regular ones.
Re: Pan Law
Posted: Sun Jan 21, 2024 4:38 pm
by utdgrant
ColinP wrote: ↑Sun Jan 21, 2024 4:26 pm
Many thanks for your free linear crossfader module, Grant!
Just looking at how CA handle pan law and it seems their 6-input Stereo Mixer and Panner modules behave in the same odd way as their Crossfader module.
When in the center each channel is at 100% and if we pan in either direction one channel remains at 100% with just the other channel reducing.
You'd expect both channels to be -3 dB, -4.5 dB or - 6 dB down in the center position if the pan law was one of the regular ones.
Cheers, Colin.
I've made an enhancement to it, so that it can also be used as a Linear Panner. And yes, I observed the same strange 100%:100% behaviour in the CA Panner module.
- CACrossfadeAndPanner.png (5.13 KiB) Viewed 32067 times
I'm currently beta-testing the enhanced Linear Crossfade, and should be sumitting it tonight or tomorrow. The left hand output is the same as the current single output. The right hand output offers the complementary mix of the two input signals. I'm not sure if the diagrams at the bottom aid understanding or muddy the waters even further!
- FrontPanelV1.1.jpg (44.39 KiB) Viewed 32067 times
Re: Pan Law
Posted: Sun Jan 21, 2024 4:57 pm
by ColinP
Nice one, Grant.
utdgrant wrote: ↑Sun Jan 21, 2024 4:38 pm
I'm not sure if the diagrams at the bottom aid understanding or muddy the waters even further!
I faced pretty much the same ergonomic challenge. I ended up presenting the VCAs as raw pairs with just normalled mono input. Then users can wire them up as panners, crossfaders or independent VCAs.
- CustomIO.png (283.17 KiB) Viewed 32050 times
Re: Pan Law
Posted: Sun Jan 21, 2024 9:37 pm
by borkman
That's the default behavior of the Mix knob on Full Thickness Tear. I've also seen it in various VSTs. For mixing wet/dry in some effects, I think its a good compromise when you don't want individual wet/dry knobs. For general crossfading, and especially for panning, it seems an interesting choice to me, but I've made (and will continue to make) interesting choices, so I will refrain from judging.