5.1 surround panning angle problem

Question

I'm having a problem getting Logic Pro to pan correctly in 5.1 surround, and the problem seems to be with the way Logic is programmed to handle the "Surround Angle" (i.e. the azimuth angle of the sound's virtual location) as it crosses the Y axis in front of the listener. I have created a soundpath (i.e., a continuously changing virtual location) for a track by moving the surround panner with the mouse and saving the automation. Here's what I'm experiencing. As the sound pans from right to left (somewhat to the front of center), the automation data for "Surround Angle" goes from, for example, 8 to 5 to 2 to 359 to 356 etc. (because the angle is going from the right of the center axis to the left of it); however, instead of the program being intelligent enough to interpolate from 2 down to 0 and then from 360 down to 359 (which would create a smooth fade across the center axis), it seems to be interpolating UP from 2 to 359, creating an obnoxious audible "swoop" as the sound very quickly pans hard to the right and makes a complete circle back to the front center. Is this just a hopeless flaw of the way the programmers have implemented the intensity panning based on angle data, or is there a known way to get the panning to act smoothly?

Answer 1

Thanks for the suggestion. When viewing the auitomation data (which by the way is not truly MIDI, it's numbers from 0 to 359 representing degrees from front center, numbers that I created by dragging on the surround panner with the mouse), it's clear that the angle is being represented as 0 degrees and upward as it moves clockwise from the front center, and 359 degrees and downward as it moves counter-clockwise from front center. The problem occurs when the sound's location moves ACROSS that front center axis (a perfectly normal and logical thing for it to do to create the illusion of left-right or right-left movement), because the automation data leaps from one extreme to another (such as from 2 to 359 in my example). Well, as Logic sets the intensity of each channel sample-by-sample as it pans, it naturally has to interpolate between angle values. But Logic is apparently programmed to do that interpolation directly EVEN WHEN the panning value goes from extremely low to extremely high or vice versa. So, instead of Logic "realizing" that a change of paning angle from 2 to 359 should be interpreted as a slight move across the front center axis, it apparently interpolates UPWARD from 2 to 359, creating a wild clockwise circular swoop. So, ...
I can't see how I can fool Logic into not doing that, and "smoothing" the transistion with additional data doesn't really solve the problem if it's occuring internally in Logic's samplewise interpolation scheme.