r3 polyphony

shaun.m.r · Post by **shaun.m.r** » Mon Mar 19, 2012 5:50 pm

using only one timbre with the unison function (page24) switched off ,how many voices are used when
(example 1) i press one note
(example 2) i press three notes at once
please be aware that on page 27 of the user manual it indicates 'the maximum number of playable voices may change depending on the combination of 'wave' and 'osc1mod' settings'.
can you tell me what the combination is and how it affects the number of voices used in the examples i have requested.
are there any other factors that affect polyphony if so how does it affect the number of voices used in the examples?.

HardSync · Post by **HardSync** » Tue Mar 20, 2012 9:43 pm

We had a discussion about some of this on a different forum here: http://www.karma-lab.com/forum/showthread.php?t=18227

Also, it's impossible to answer your question's examples because you haven't specified what the voice stack is. If you set it to two, then you will use at least 2 voices per key press. If you set it to 3, then you will use at least 3 voices per key press. And 4 will use 4 voices with one key press.

(Edit: I think I misread your question. If you have the unison switched off, then you will use one voice per key press, unless you have the wave shaper stuff turned on.)

As for the OSC mod and stuff: Firstly, there is a difference between the Unison function setting, and the OSC1 Mod Unison. The Unison function mentioned on p24 will decrease the amount of available voices (or keys you can play at once), while the OSC1 Mod Unison doesn't affect the number of voices.

As for things that can affect the voice count, some of the Waveshaper stuff needs to use a voice. Certainly the SubOsc ones will use up a voice...

The easiest way to find out if something decreases the polyphony is to hold down the damper pedal and start adding notes one by one, counting as you go. When the first note you played is cut off, then you know how many voices that timbre has used with the settings you have selected.

shaun.m.r · Post by **shaun.m.r** » Thu Mar 22, 2012 2:14 pm

i do not have a damper pedal . also the question i asked does not relate to
the osc1 mod unison but to the fact that the manual states there is a
combination of wave and osc1 mod settings that affects polyphony(page27).
as for the waveshaper the manual does not indicate polyphony loss in that area.
out of the entire manual polyphony loss only occurs on page 24 and page 27.
can anyone answer my original questions?

Re-Member · Post by **Re-Member** » Thu Mar 22, 2012 4:22 pm

I hate to sound like a wise guy and state the obvious here, but no one needs a damper pedal to test the polyphony. You only need a maximum of eight fingers.

Also, I seem to remember that it was the VPM feature that ate up an extra voice. Don't have access to my synth at this time to comfirm, but here's a statement regarding VPM from the manual: "Oscillator 1 relies on its own sine wave generator, so unlike Cross Modulation, oscillator 2 has no effect on VPM."

Since VPM actually generates an additional oscillator to act as a modulator, it seems like that would be taking up an extra voice. The SubOsc feature of the DRIVE/WS section shouldn't eat up polyphony since there's nothing stated on that page, but again, can't test this myself yet. It seems likely that the SubOsc feature is PCM based whereas the VPM Osc is actually generated.

HardSync · Post by **HardSync** » Thu Mar 22, 2012 4:22 pm

You don't need a damper pedal. The R3 has 8 voices. You (probably) have 10 fingers, and even if you don't then you could increase the decay, sustain and release times to let the notes ring out. You could use a book on the keys, your toes, or a shoe, or maybe a cat (keyboard cat, perhaps?), get a friend to help you. I'm not seeing any difficulty in you being able to test it here... really not seeing any trouble at all.

I can only guess at the answers to your questions. I don't know if guessing will be good enough for you, but I'll give it a go anyway. If you aren't satisfied with the below, then I would recommend e-mailing Korg and asking them for a definitive response.

Combinations of things that could use up additional voices:

1. VPM uses a sine wave harmonic on oscillator 1 to do its thing. That could probably take up a voice. I say "could" and "probably" because I don't know for certain and I can't be bothered to test it.

2. As mentioned before the Wave Shape SubOsc settings will use a voice.

example 1 answer: 1 voice if not using VPM or Wave Shape Sub Oscillators, 2 voices if using the sub osc, possibly 3 voices if using the sub osc and VPM.*

example 2 answer: 3 voices if not using VPM or Wave Shape Sub Oscillators, 4 voices if using the sub osc, possibly 5 voices if using the sub osc and VPM.*

As for the osc1 mod unison thing, I was telling you that so you would understand that it does NOT affect polyphony like the other Unison, just in case you didn't know, because I have no idea what you do or don't know. But hey, that's just me -- this random guy on the internet -- trying to help you and others by at least attempting to answer your questions. It was sad to see this topic sitting there unanswered, so I tried... and um... never mind.

HardSync · Post by **HardSync** » Thu Mar 22, 2012 4:25 pm

Re-member, we posted at the same time, pretty much saying the same thing... nice.

Synthoid · Post by **Synthoid** » Thu Mar 22, 2012 4:28 pm

Fascinating.

shaun.m.r · Post by **shaun.m.r** » Thu Mar 22, 2012 5:31 pm

thanks for trying to help,the vpm solution sounds good,not sure about the waveshaper.I also have a radias and in the manual it says that the waveshaper uses extra polyphony,so i hope you are wrong about the r3 because it does not mention this loss in the r3 manual.
i have tried testing with fingers but cannot tell when first note cuts off,the thing is i need to work it out without fingers because my next problem is working out polyphony loss on the radias and i do not have 24 fingers.

HardSync · Post by **HardSync** » Thu Mar 22, 2012 6:32 pm

I've tested it on an EXB-Radias. None of that stuff uses any additional polyphony. Not any of the Wave Shape stuff, not even VPM, nor Cross Mod, nor Waveform. I tried every possible configuration and combination, with the all of the various wave shapes turn on and off. I can play a full 24 notes all the time. Every time. The only things I did not try was modulating timbres with LFOs or using a mod sequencer.

It's not an R3, though.

If you want to test this with fingers only, you need to A) set the amp envelop velocity sensitivity to max, and B) work out which notes the R3 gives priority to before cutting off notes when keys are held down (high, low, first played, it will be one of these three) -- when I use a damper pedal the first note played will cut off.

After you've worked that out, play your first note hard at max velocity, and play the remaining notes really softly, so you can only just barely hear them. When the loud note goes away, you will definitely hear it.

X-Trade · Post by **X-Trade** » Thu Mar 22, 2012 8:19 pm

In the Radias, Comb filter reduces polyphony from 24 to 16. Sub oscillator wave shaper settings take up an extra voice per voice.
I'm not aware of any Osc1/mod settings that us up additional voices.

shaun.m.r · Post by **shaun.m.r** » Fri Mar 23, 2012 2:26 pm

thanks again, i'm trying to get korg to give me the facts but they do not
seem very quick to answer.

shaun.m.r · Post by **shaun.m.r** » Sat Mar 24, 2012 6:59 pm

for anyone that is intererested,korg replied to me saying,
A single timbre with unison switched off provides the maximum of 8 voices.
also korg have had difficulty finding any combination of 'wave' and 'osc1 mod' that affects polyphony.(although it does say on page 27 of user manual that there is a combination that affects polyphony).

Re-Member · Post by **Re-Member** » Sun Mar 25, 2012 12:10 am

Korg is indeed correct. I tried every OSC 1 and OSC 2 mod setting, plus all the Drive/WS effects and none of them used an extra voice. The only thing that eats up extra voices is having Voice Mode set to Layer on P01 (since both timbres are stacked and triggered simultaneously) and/or having two or more voices for the Unison effect on P02.

Again, only took me eight fingers to figure this out, haha.

Re-Member · Post by **Re-Member** » Sun Mar 25, 2012 1:38 am

Forgot to test the comb filter. Just did so and it actually does drop the polyphony down by one note.

shaun.m.r · Post by **shaun.m.r** » Tue Mar 27, 2012 11:58 am

AAAAAAAgh what is going on? how can the manuals be so wrong,even korg has difficulty finding its own polyphony loss!
now you tell me the comb filter uses extra polyphony? could it be possible that you filtered out a note? cutoff a note perhaps?