Tip: Fantom Polyphony and Voice Assign Mode

[xp30]

Tip: If you are concerned about running out of polyphony while playing phrases with hold pedal, and you are not sure what Voice Assign Mode does, then try setting Voice Assign Mode to SINGLE.

There have been many discussions about polyphony on the Roland Fantom (and related Zen-Core hardware). I think Voice Assign Mode is the most important knob to minimize polyphony issues for typical playing with hold pedal, and interestlingly, it also seems to be the least discussed.

Let's make an example. The following scene is an initialized scene with a clip in the first zone. I have turned off the reverb to make it easier to hear everything:

study#82.zip

(2.9 KiB) Downloaded 114 times

Let's configure the first 3 zones with:

• Zone 1: SN-AP:EXSN03:0001:Classic Piano with level 120.
• Zone 2: Z-Core:PR-E:0387:RD-1000 with level 60.
• Zone 3: Z-Core:PR-B:0162:Soft Pad 1 with level 40.

Then change Voice Assign Mode to FULL for the first 3 zones, copy the clip from zone 1 to zone 2 and zone 3, and then start all 3 clips. This sounds as follows:

study-82-vam-full.ogg

(284.58 KiB) Downloaded 163 times

I can clearly hear that many notes are cut off.

Now let's change Voice Assign Mode to SINGLE for the first 3 zones, and this time it sounds as follows:

study-82-vam-single.ogg

(283.77 KiB) Downloaded 176 times

There are still a few notes that are audibly cut off. However, I think it is good enough, so that most audience members would not notice it during a live performance.

Let's go one step further: As mentioned previously (https://forums.rolandclan.com/viewtopic ... 66&t=69417), the tones Z-Core:PR-E:0387:RD-1000 and Z-Core:PR-B:0162:Soft Pad 1 are wasting polyphony, because they both use 2 hard-panned mono partials instead of 1 stereo partial. After making the optimization, it sounds as follows:

study-82-vam-single-optimized.ogg

(282.78 KiB) Downloaded 281 times

Now I have about 50% more polyphony than I need for this clip, or put differently, I could add another 2 layers without any notes audibly cutting off.

The Voice Assign Mode also exists on other Zen-Core hardware products (MC-101, MC-707, MV-1, AX-Edge, Juno-X, Jupiter-X, Jupiter-Xm), and if you use them in a similar way (i.e. repeatedly triggering the same keys during a hold pedal phrase), then consider setting Voice Assign Mode to SINGLE on those products.

See also https://forums.rolandclan.com/viewtopic ... 66&t=69412 for more information on Voice Assign Mode.

[FGM]

Thank you very much.

I was standing by the guitar player, seven other players around.

All was great, sounding beautifully.

When the music was over he approached me and yelled: I cannot belive no one noticed one of my strings broke.

😀😀😀😀
🎹🎹🎹🎹

[xp30]

All was great, sounding beautifully.

All 3 of them? I wonder how many people notice that a few notes were cut off in the second example.

[FGM]

All was great, sounding beautifully.

All 3 of them? I wonder how many people notice that a few notes were cut off in the second example.

Hard to ascertain but for the composer.

At times I ask my friends and family which version of a little part of a song they like their most, playing barely for ten seconds to show OBVIOUS differences, and the answer I get is they don't notice anything, be it a change of octave, a different tone providing tension, added arpeggio...Pffff.

I myself discovered a horse cavalry was missing when playing a song of mind, composed weeks before, after realizing the wireless transmitter, advertised and sold as stereo, was in fact duplicating the left channel, and the horses had been placed on the right one. But it took hours...I simply went questioning a part that was lacking some harmonic support without remembering those missing notes, until later on...

😀😀😀😀
🎹🎹🎹🎹

[xp30]

Hard to ascertain but for the composer.

At times I ask my friends and family which version of a little part of a song they like their most, playing barely for ten seconds to show OBVIOUS differences, and the answer I get is they don't notice anything, be it a change of octave, a different tone providing tension, added arpeggio...Pffff.

I understand what you are saying. However, this demo is not about the composition. Its main purpose is to demonstrate the effects of Voice Assign Mode on perceived polyphony (i.e. notes seemingly cutting off). I wonder how well the demo achieves that, and if people notice any other differences.

[FGM]

I know, I know...

Just a comparison dealing with human perception.

Another one ? Remember that actor being killed while the attendees still yet laughing because it was funny and, they thought, part of the comedy.

😀😀😀😀
🎹🎹🎹🎹

[shaneblyth]

Fantom O series rather than the big brother is their a similar voice assign mode as I searched manual didn't see one not that the manual is complete if so where is voice assign. I only just got the O7 so not up on it much i noticed you in another posted talked about half damper settings is that in the O series as well again searching the pdf manual I can't find where these settings are if they are in there.. cheers..

[xp30]

Fantom O series rather than the big brother is their a similar voice assign mode as I searched manual didn't see one not that the manual is complete if so where is voice assign. I only just got the O7 so not up on it much i noticed you in another posted talked about half damper settings is that in the O series as well again searching the pdf manual I can't find where these settings are if they are in there.. cheers..

Yes, both features also exist on the Fatom-0. You can find documentation of almost all parameters in the Parameter Guide. Voice assign mode is a zone parameter and documented on page 16. Redamper is a Zen-Core partial parameter and documented on page 34. Note that the simplest way for turning off the redamper feature is to switch envelope mode from NO-SUS to SUSTAIN.

[xp30]

I updated today from version 3.01 to 3.02. Before the update, I verified that the notes were still cutting off with voice-assign-mode set to LIMIT for FULL. After the update, I can no longer reproduce the issue. It seems that with version 3.02, it is no longer necessary to tweak settings to avoid such polyphony issues.

[FGM]

I updated today from version 3.01 to 3.02. Before the update, I verified that the notes were still cutting off with voice-assign-mode set to LIMIT for FULL. After the update, I can no longer reproduce the issue. It seems that with version 3.02, it is no longer necessary to tweak settings to avoid such polyphony issues.

Many thanks.

Then I will update...

😀😀😀😀
🎹🎹🎹🎹