Korg Forums

Hi -

I want to see if it is possible, through OASYS software, to emulate a kind of pseudo polyphonic aftertouch capability on OASYS - or at least note or voice-specific modultaion control of EXi Synth engines.

As many of you know, I harp on about how great polyphonic aftertouch is on the likes of the CS80 and Prophet T8, and alas OASYS (and Kronos) do not have that feature.

What they do offer however is polyphonic LFO's, meaning that for each of, say, the 52 voices of a MOD-7 patch, each has its own LFO. This is different to many synthesizers that share one LFO across all voices.

I've exploited this advantage in some MOD-7 programs by making the LFO rate velocity sensitive, so that if you hit, for example, two keys at different velocities then the LFO rate will be different for each key. It's a wonderfully expressive effect.

However, the effect is only velocity sensitive; and I wonder if it would be possible to combine modultaion controller sources (using the AMS mixers) in unusual ways to provide for the use of polyphonic LFO features that can be exploited independentely in other ways other than note velocity? Ideas that come to mind (but which I haven't given thought to yet due to time constraint and hence throwing it open here) might include:

- pseudo polyphonic aftertouch on other features other than traditional LFO effects (combining say note number with channel aftertouch or channel aftertocuh and some other weird LFO feature such as a kind of DC Bias then applied to, say, filter cutoff?)

- combining polyphonic LFO settings with other controllers such as the vector joystick, channel aftertouch

- wavesequencing features affected by polyphonic LFO's....

- AMS mixing several LFO's affected by different amounts by note velocity

I'm not really too sure what I precisely mean as its been a long while since I read over these features in the OASYS manual; but I am convinced that there are unique performance features to be exploited from the OASYS in these regards.


My motivations for this come from my recent delving into my Yamaha DX1 properly for the first time, only to discover the amazing capability of this instrument to affect, through polyphonic aftertouch, the levels of individual operators in an FM algorithm, with each operator having its own polyphonic aftertocuh sensitivity. This means that, on a per note basis, you can, for example introduce new aspects to a sound not heard on other notes, or even new sounds not heard on other notes, though polyphonic aftertouch. For example, using an algorithm with 2 carriers, you might make carrier one form the basis os a piano sound, and carrier 2 the basis of a bell sound but only heard on notes where the level of carrier two only sounds when aftertouch is applied. Of course you can apply the same principle to modulator operators also, allowing for any type of modulation to be introduced and controlled on a per-note basis. It's an awesome feature and is what makes the DX1 so incredible - basically turing a static FM sound into a world of performance possibilities.

And of course OASYS is filled with similar possibilities, especially though its AMS mixers, and if anyone has put any thought into exploiting these for performance possibilities I'd love to hear about them.

I indicated over a year ago that I'd try to produce more MOD-7 programs with advanced features like this but haven't had the time. But as and when I exploit these features I'll be happy to share them.

Kevin.

-

Kevin, I've not enough knowledge of how to achieve what you were talking about with the Oasys. But I want to remark, that a thread like this one is IMHO much closer to what this forum and the Oasys section here is (should be) about than many other posts I've read during the last weeks (including some of mine ). Thanx for starting this one and I'll follow it with great interest!

+1

....but Charlie, seriously, I dont mind your complaining at all

+1

I admit I've been complicit in taking the O forum 'off-piste' but I miss the learning resource it once was.

Thanks KN for taking the lead... wish I could help but it's beyond my knowledge but I will be following with great interest; such advanced techniques fascinate me!

+1, so we are all here.

Thank you for the idea. (I understand but I don't have the time to try soon.)

Cheers -

Yes - there are surely a myriad of nuggets of gold under the hood of the OASYS for us all to uncover over time.

I am pursuing some programming along these lines and if over the months any interesting programs arise I'll be sure to post them.

And along exactly these lines Sharp has just posted that really cool way to use OASYS Combi mode as a draw-bar organ - a real 'lateral thinking' idea from him - can't wait to try it out.

Cheers!
Kevin.


PS - it's a time like this we also need an old OASYS forum member 'sirwombatcombat' back - not sure if you remember him but he was always full of intriguing ideas about OASYS (and lists a hammer and anvil among his instruments!!)

If there's channel AT, and capability to play a 16 channel combi, then a pseudo poly-AT could be gotten (for ~3 octaves or so) by mapping adjacent keys to a channel(s) spread over a short length of keyboard.

1 = C-C#-D
2 = D#-E
3 = F-F#
4 = G-G#
5 = A-A#-B
... 16

Using this scheme you could get poly-AT for notes such that most triad chords would not have 2 notes of the chord on the same channel. The more open the chord the less likely any AT overlap.

Zis 'elp?

BB

billbaker wrote:If there's channel AT, and capability to play a 16 channel combi, then a pseudo poly-AT could be gotten (for ~3 octaves or so) by mapping adjacent keys to a channel(s) spread over a short length of keyboard.

1 = C-C#-D
2 = D#-E
3 = F-F#
4 = G-G#
5 = A-A#-B
... 16

Using this scheme you could get poly-AT for notes such that most triad chords would not have 2 notes of the chord on the same channel. The more open the chord the less likely any AT overlap.

Zis 'elp?

BB

You're missing the point that the keybed itself only has a single monophonic pressure sensitive strip - not a sensor for each key. Key zones won't have any effect on aftertouch, all timbres will respond the same.

@Kevin: yeah - I remember combatwombat and haven't seen him around here for a long time. His bell-drone was amazing!

OK let me try again....

Keybed is mono AT.

Point of exercise is to generate an AT message imitative of poly AT.

You could, I suppose modulate, using the ribbon controller, with the ribbon sensitivity tied somehow to the note number -- then say you were playing an open C chord C-G-E, you could "find" the triad on the ribbon and effect a single note as selected by the ribbon.

But I don't know how you'd provide a mod amount or intensity without adding a 3rd controller (CC pedal?) into the performance equation.

---------------

This is about the point I say look for a vintage synth with multi AT keybed and just do it via midi. Might be more cost effective in the long run + more sounds + not a compromise + no ergonomic / performance learning curve.

BB

how about AT x Vel?

its not a great solution but you're getting different AT response from different keys. I don't know if you can do that with AMS mixers but I imagine so.

Is that Velocity +/- mono AT amount?

Seems to me that the Mono AT stumbling block is still there if more than 2 notes are played.

BB

X-trade -

I had though of that and have tried it - and I think it worked (its about a month since I tried that out but I think I got it to work). Remember BB - the overall modulation effect comes form the AMS mixer which is a combination of AT and Velocity - so it's note sensitive because velocity is note sensitive.

I must check this out - what if velocity sensitivity is not actually applied to any specific synth paramters on its own, but then Velocity x AT applied as a modulator say to filter, vibrato or vibrato speed and so on? You could apply this to, say a sawtooth synth patch, where things like filter broadening and LFO effects could be controlled on a note by note basis simply by hitting that note harder when you know you want to hold it and apply some expression. In this instance, the value/range of AT and Velocity sent into the AMS mixer is important - you'll always get a little AT even on other held notes, but you'll get more modulation effect from notes hit harder and then leaned into for aftertouch.

That might work (??)

Kevin.

Kevin -

Yeah, but...

As soon as you leaned into it ALL notes would be affected (Mono AT) - some might be affected MORE, but the AT "lean" would be a blanket modulation.

How would you keep the modulation damped down to the musically useful level - such that almost any AT would not bring effect on ALL notes to maximum?

My usual velocity is in the 90 to 110 range with ff being 120-127 - could be more but keys max at 127. Assuming fairly light playing - two notes - one v=80 (mp), 1 v=110 (mf) ... AT = 0 would have a pretty noticeable difference to start.

Lean, AT = X ? [X . 80 = ?] [X . 110=?] == wouldn't it be close to max on both right away?

--------------

The solution using velo would seem to me to be to ignore AT altogether and tie some filter or other effect straight to velo.

NOT using velo for amp control (i.e., always max amplitude as in a true organ) would make the effect easier to use, but you'd clearly need to make radical adjustments in playing style if you were trying to emphasize an inside voice as opposed to a right hand lead over comp chords.

BB

billbaker wrote:Kevin -

Yeah, but...

As soon as you leaned into it ALL notes would be affected (Mono AT) - some might be affected MORE, but the AT "lean" would be a blanket modulation.

How would you keep the modulation damped down to the musically useful level - such that almost any AT would not bring effect on ALL notes to maximum?

My usual velocity is in the 90 to 110 range with ff being 120-127 - could be more but keys max at 127. Assuming fairly light playing - two notes - one v=80 (mp), 1 v=110 (mf) ... AT = 0 would have a pretty noticeable difference to start.

Lean, AT = X ? [X . 80 = ?] [X . 110=?] == wouldn't it be close to max on both right away?

--------------

The solution using velo would seem to me to be to ignore AT altogether and tie some filter or other effect straight to velo.

NOT using velo for amp control (i.e., always max amplitude as in a true organ) would make the effect easier to use, but you'd clearly need to make radical adjustments in playing style if you were trying to emphasize an inside voice as opposed to a right hand lead over comp chords.

BB

BB you're correct to a large extent but I think we can extract more subtlety out of it with a little experimentation among the scaling at the LFO , the AMS and the destination being modulated. In essence, high velocity need not mean high modulation - but rater we need to find the 'sweet spot' across the difference between velocity levels of say 80 - 120

I'm not saying I'm right, I suspect I encountered the very issue you raise and that's why i can't quite remember how successful my efforts were (done in a hurry by the way).

the advantage of introducing the effect from aftertouch rather than velocity of course is that you control the modulation after the attack of the note and according to pressure, especially effective on prolonged notes.

But all things told - this approach is probably more effort than it's worth, but I still think it's worth exploring - it gives you a 'feel' for the synth engine LFO's and ams mixers

Kevin.