Pan Spread Issues

[zoso335]

Hey everyone, I recently discovered that the pan spread on my Rev2 doesn't fully function and wanted to see if anyone knew of a remedy. I also tested my Prophet 12 and Prophet X and it doesn't fully function for those either!

With pan spread set to max on a basic program, 1 or more voices are closer to center than L or R. I tested with headphones plugged into the synths.

I calibrated the oscillators/filters for all but that didn't help. Does anyone know how this can be fixed?

[creativespiral]

Hmm... just tested on my Rev2-16 Voice via main outs.   I do see what you're seeing. 

Pan Mode: Alternate
Pan Spread:  Maximum

Hitting the same key over and over, the L/R spread does vary... there must be some other calculations going on...  by the looks of it, seems like it's a repeating pattern based on voice number.   Maybe there is a +/- pan offset associated with each voice number? 

[zoso335]

So I contacted Sequential and they said this is how pan spread normally operates: progressively getting closer to center.

If anyone wants the voices to continuously alternate between L and R, I discovered you can do this in the mod matrix by having DC modulate Pan at maximum amount with the pan mode set to alternate.

[creativespiral]

Hmm.   Makes sense. 

I have wondered in the past about the scaling of DC amounts to Pan destination.   In Fixed pan mode, sending -32 or +32 pan amount fully pans the layer hard left or hard right... anything beyond that has no real effect.   I use this method often for creating binaural patches / wide stereo fields. 

This may be the reason that they used that scaling factor... so in Alternate mode, you could use a mod slot to "force hard panning" in an alternate fashion by dialing in more modulation than from the Pan Spread potentiometer alone.

[AlainHubert]

So I contacted Sequential and they said this is how pan spread normally operates: progressively getting closer to center.

If anyone wants the voices to continuously alternate between L and R, I discovered you can do this in the mod matrix by having DC modulate Pan at maximum amount with the pan mode set to alternate.

Thanks for this. I've always wanted to do exactly that, but never thought of using DC in the Mod Matrix to do it. I wanted to recreate the Oberheim OB8 (?) sound at the start of Natural Science song by Canadian rock group Rush. Unfortunately the random waveform of the LFO is not unified (a single randomized value for all voices) to modulate the filter cutoff, so I can't replicate it exactly. But at least the alternate panning is there.

[Bing]

Another Panning issue...I set the LFO to a square wave adjusting the pan.  Anytime it switches from left to right or vice versa there's an annoying high pitch click.  I feel like I read about this problem somewhere else but I can't find it.  Was this a known issue with a solution?

[creativespiral]

Another Panning issue...I set the LFO to a square wave adjusting the pan.  Anytime it switches from left to right or vice versa there's an annoying high pitch click.  I feel like I read about this problem somewhere else but I can't find it.  Was this a known issue with a solution?

Hmm... not in front of my Rev currently, but I know this is an issue that many synths face when routing squares LFOs to pan.   I have a Reaktor synth that I'm building where this type of issue caused hours of debugging.  It needed to have a smoothing algorithm between LFO and Pan mod reception.

I'm assuming you're in FIXED pan mode in the program settings?    You might try using a Sine LFO, with very high amount... that could possibly get you a "square like" hard pan sound, with some smoothing... I'm thinking high enough amount would cause the top and bottom of sine wave to clip, but the transition would still have some curve/smoothness.   Or alternatively, you could try using Lane 1 of gated sequencer mapped to Pan modulation, and use Lane 2 of the sequencer to Slew the first lane.   (you would probably need to use a DC offset to allow for panning both directions with gated seq)    Again, not currently in front of my board, so can't test, but just thinking these might be solutions to get that Square LFO > Pan type of sound... which can be awesome when its synchronized to a beat and doesn't have artifacts.

[Bing]

I'm relatively new to synths in general and had to look up some of that jargon.  What I was specifically trying to do was get some sort of Rhodes hard panning effect.  The clicking sound only comes out when my cutoff is relatively low.  I think the higher frequencies kinda mask it a little bit.  As of now, the LFO pan as a triangle wave sounds close enough. 

I was in alternate pan mode, but switching to fixed didn't make a difference.  Not sure what you mean by sine LFO, I don't see that waveshape on the rev2 panel.  Is it somewhere else?

As for gated/cv stuff, I have not yet scratched the surface of that in the synth world so I'm not entirely sure what it means.

[creativespiral]

I gave some wrong info in the previous post... that's what I get for trying to debug without having the board in front of me! Yeah, there is no sine lfo on the rev, so you'd have to use the triangle in that example.

There are a few things you may want to note about LFO output values:

The Square, Sawtooth, and Reverse Sawtooth output uni-polar values:  ie:  0 to +127

The Triangle and Random output bi-polar values: ie: -127 to +127

So if you're routing to Pan destination in fixed mode, and using the square shape, you'll need to have an offset via the mod matrix, so the final output values will range from say:  -64 to +64


I just did a test with the following settings:

Program Pan Mode: 
Fixed

LFO 1:
Square
Rate: To taste
Destination:  Pan
Amount:  127 (max)

Mod Slot 1:
Source:  DC
Destination: Pan
Amount: -64

These above settings give a very musical sounding lfo pan, with minimal artifacts...  of course, once you crank up the rate and get near audio rate lfo frequency, things start to get mangled a bit.. but that's normal behavior.    If you keep the rate down lower or beat sync'd to 1/8 or 1/16 notes, it sounds really tight.

Alternatively, using the triangle LFO in fixed mode, you wouldn't have to use an offset in the mod matrix, since the triangle outputs bi-polar values:  (-) and (+).   Sounds like this is giving you a sound you like already.

One other thing to note:   You can route LFOs to Destination in the Mod Matrix too, in addition to the direct routing of LFOs to destinations.   When you route them using the mod matrix slots, the output value is 4x greater.   I posted some more information in the below thread with some technical details:
https://forum.sequential.com/index.php/topic,3203.0.html

[Bing]

Thanks so much for the info!  I'll mess around with it tomorrow.

[Bing]

Tried it with the exact specs you posted but I'm still getting the same results.  I see your logic though.  I'll mess around and see if I can find any other solutions.