StereoScope Graphing Error

[kcisANDderit]

i've been meaning to post about this. the StereoScope graph is off between center and 100% Left/Right (possibly beyond too)

A signal panned 50% is graphed further than 50%. 50% on the graph is actually 38%

[Blue Cat Audio]

Hi,

the value for the position actually depends on the panning law, and the one we have chosen for the gain plug-ins is not the same as for the StereoScope Series. The panning law to chose the center in the widening gain plug-in is linear, whereas for the Stereoscope series it is close to a cosine law, which is closer to our perception of the stereo field.

We will probably propose several panning laws in the future, in order to improve the coherence of our products.

[kcisANDderit]

I tried a few other panning plugs and the behavior was the same, they're all linear as would be expected

so we perceive 50% left to be further left than reality (i've never read about stereo perception)

[Hayate Yagami]

StereoScope:
L - R = 7dB --> 50%L
R - L = 7dB --> 50%R

Generally:
L - R = 6dB --> 50%L
R - L = 6dB --> 50%R

-6dB = 0.1^(6dB / 20) = 0.501187... = about 50%

The panning law influences the envelope-height, but it doesn't influence the envelope-position(50%L / C / 50%R etc).
I think so.

I have been used to StereoScope already.
All meters have a habit.

[kcisANDderit]

The panning law influences the envelope-height, but it doesn't influence the envelope-position

it does influence the envelope-position. I tried it with 0, -3 and -6

[Hayate Yagami]

The panning law influences the envelope-height, but it doesn't influence the envelope-position

it does influence the envelope-position. I tried it with 0, -3 and -6

It means the difference between the R-ch and L-ch level influences the envelope-position.
However any panning law,
L(level) - R(level) = 6dB --> 50%L
R(level) - L(level) = 6dB --> 50%R
I think this is the ideal.

[Blue Cat Audio]

Anyway the actual perception of the human hear in the stereo field is much more complex: it depends on many parameters, such as the monitoring system (headphones or speakers as well as the position of speakers), the spectral content and delay between left and right.

So it means that the scales for measurements have to be arbitrary in some way, just like the pan laws. One thing we could add to the product to improve this is to offer several types of scales so that the measurement would fit the particular method you are used to.

[kcisANDderit]

well imo, since you can't account for everyone's perception, the simplest and best solution is to make it linear. if someone turns the panpot 50% then the scope should read around 50%

[Blue Cat Audio]

That's a good point. We'll take your remark into account when we think about the next update.

[kcisANDderit]

well imo, since you can't account for everyone's perception, the simplest and best solution is to make it linear. if someone turns the panpot 50% then the scope should read around 50%

in v1.9 cosine gives me the above, not linear

is that as it should be?

just asking cuz I really don't understand the math anyway, but if the plugin doesn't display the options as you intended then i'm letting you know

[Blue Cat Audio]

It depends on the settings in your host and the way they decided to call pan laws . Typically in Sonar the default pan law is sine/cosine, and that's a pretty common setting, so "Cosine" is indeed the way to go!

BTW be aware that the new scaling will be 100% in sync with your daw for mono signals only, since for stereo signals the perceived position depends on how the sound is spread between channels.

Hope this helps!

[kcisANDderit]

got it. thx