Slots A and B Polyphony

[Mr_-G-]

Hi,
Just to be sure, if you use both slots A and B at the same time, does this reduce the max polyphony?
Thanks,

-G-

[mjbrands]

That is a very good question, I'd like to know the answer myself. The Nord literature just mentions 40-60 notes polyphony, it doesn't mention if that is per slot or total and what has an impact on that total (maybe features like string resonance and long release).

[maxpiano]

40-60 is for piano only, organ is full polyphony and synth is 18 notes.
Anyway the question is a good one, at least for Pian & synth: has anyone tested this? (for synth should be easy, take a layered synth sound and try building a 10 notes chord one ote at a time then "unbuild" it: if the last note won't play then the polyphony is 18 total = 9 when layered)

[mjbrands]

I'm convinced it is 18 in total for the synths section, where the voices are dynamically allocated across the two slots (that's also how the Wave works). Easy enough to test, as you say.

But what about the piano section: it is probably 40-60 voices (notes) for the section, with voices dynamically allocated to both slots with a smart note-stealing algorithm to hide hitting the polyphony limit. How would you test this. One thing I can think of is start playing (MIDI file, so it is reproducible?) and recording the resulting sound. You might be able to identify the individual notes based on their frequency.

I'm probably making things more difficult than they need to be; someone with good ears would likely be all it takes.

[Mr_-G-]

Thanks for the comments. I also thought that they might be dynamically allocated to each slot. However this does not seem straightforward to test.
For instance I have a synth patch using banks A and B. When I play the last note on the left (A) and press the sustain pedal, then I seem to be able to play more than 18 notes starting from the far right, and the first sustained A is still not muted, while all the rest also appeared to be still sounding. I must say that this was surprisingly better than what I was expecting.
So I can only guess that other notes probably get muted, but not in the original order they were played?

[maxpiano]

I made the test, playing a 2 hands (10 notes) C# diminished chord on the preset "Lisa sound", building the cchord from the lowest note (C#3) and then releasing the notes one by one (starting from the highest)

When the last two notes/fingers are left (the lowest C# and E) you will notice that only one note is sounding: this shows that 18 is the total (A+B) polyphony so when slot A and B are layered you get 9 notes max. NOte: the last note sounding is not the E3 as I expected but the C#3 , this must be the result of the smart note stealing algorithm

@mjbrands: you can do the same to test the Piano, but you need some help because you need to play at least 21 notes to go over the (theoretical) limit of 40/2=20; or (as you said) create a MIDI file that does the same (21 note-on followed by 19 notes off, then check if 2 notes or 1 are still sounding)

[Mr_-G-]

Hi,
Thanks maxpiano, yes that is a good example. Test this now (same Lisa sound would do):
Play the last note (A) on the left and press the sustain. Now play lots of notes on the right end of the keyboard, still with the sustain pressed. I never get the low A voice stolen from the new ones (but some of those high new notes obviously get muted when reaching the polyphony limit). I am well pleased with this algorithm. Comparing to other (older) synths (like the Roland JX8P) the voices get muted in the order they were played (and the effect of that is sometimes awful, as you suddenly loose a root note).

[maxpiano]

Yes I agree, the smart voice allocation algorithm of the NS2 Synth is actually effective and handles well the most common playing situations.

Let's also remember when Analog synths with just 8 voices were the rule... I think NS2 synth polyphony is OK for average and intended usage of the instrument, for leads or solo sounds 9 is more than enough, for pads to be layered to a piano maybe it is better to use just 1 slot and stay 18 (even if... a real strings orchestra would seldom play 18 different notes at a time, so having lower polyphony somehow helps to get a better execution realism too)

[mon8169]

On the classic you have 2 independent synths (and other engines) so you have double polyphony. At least I always have believe it! Isn't it the same on the Stage 2?

[Mr_-G-]

According to the test above, It does not appear to be so... 9 if you use both slots.
I wonder about the pianos too. Same story?