posted 30 July 2005 11:57            

I have a sound that uses !KeyPitch. This works either with a MIDI keyboard or with a Wacom tablet.

Now there is a !Duty parameter I'd like to bind the Pen Tilt. To do that I bring up the pen windows by clicking on the pen in the VCS (the pen is grayed at this time).

If I bind the Eraser's PenTiltRadius parameter to !Duty, I can not use the Eraser PenX parameter to control the !KeyPitch no more.

I have to revert the sound to its original values in order to use it again with the eraser side to control the Pitch.

Is there a way to defin the eraser's PenX parameter to !KeyPitch ?

posted 30 July 2005 19:30            

quote:

---

Originally posted by SSC:
The eraser end controls !KeyPitch by default. You do not have to map PenX to KeyPitch. (But you could map PenX to control a pitch or frequency with the pointed end of the pen).

---

Yes, but if you map Eraser / PenZ to something then you loose the ability to control KeyPitch with Eraser / PenX.

Another strange thing happens when you do this simple test :
Get the oscillator, in Frequancy, put "!KeyPitch smoothed", and in Enveloppe, keep "!AmpLow smoothed". If you try the sound, the pitch is as expected controlled by the Eraser end. Then bring up the pen window and assign the !AmpLow parameter to Eraser / PenZ.
You loose the pitch control, and the !AmpLow doesn't seem to be smoothed. Artefacts can be heard, something like aliased enveloppe changes. It is easy to seem this distortion with the spectrum analyser.
If you bind the !AmpLow parameter to a MIDI controller, the artefacts are still there, so it isn't a pen problem.

I saw that if I put an attenuator behind the oscillator with "!AmpLow smoothed" as an argumen and use the interpolation, the artefacts disappear. The interpolation was checked in the oscillator though, and seem to interpolate the frequency only, is that right?

I would like to learn more about this question, and I am curious to know why the "smoothed" message does not smooth the enveloppe as it should !

posted 31 July 2005 11:50            

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

posted 31 July 2005 14:16            

quote:

---

Originally posted by SSC:
In the Pen Status window, the Eraser / Pen control selects which end of the pen sends keyboard events and which end of the pen sends continuous controller events. This means that when you choose Eraser and then map Pen Y, the keyboard end is switched to the Pen end of the pen.

---

So that means the keyboard is always mapped, which is good.
But how do I map velocity to the eraser end PenZ, with the keyboard active on this end?
I find it a bit weird to have a parameter "ghostly" bound to the pen, ie not showing in the pen window.

quote:

---

I just tried the Oscillator set up the way you described and I did not hear the distortion. "smoothed" makes changes at a 1 khz rate (which may be audible in certain situations). The Attenuator with Interpolation turned on is interpolating at the audio rate and should rarely be audible.

---

I was surprisedm but I can tell the Interpolation turned on doesn't absolutely nothing on my side ; which is why I had to put an Attenuator with Interpolation after the Oscillator.

I will send you wave and kyma files demonstrating the problem in another topic as soon as I grab a hand on my computer.

Thank you very much for your answers.

posted 31 July 2005 17:39            

Keyboard events are associated with one end of the pen; continuous controllers are associated with the other end. (By default, the eraser end is the Keyboard). You cannot have key events and continuous controllers on the same end of the pen. The only exception to this are the "tilt" controllers which work on both ends of the pen.

By changing the name of the hot value, you could control keyboard-like events mixed with continuous controllers. For example, if you use (!PenX * 24 nn + 60 nn) in a Frequency field you would have a 2 octave range starting at middle C that you could control from the pointy end of the pen.

"I was surprisedm but I can tell the Interpolation turned on doesn't absolutely nothing on my side ; which is why I had to put an Attenuator with Interpolation after the Oscillator."

Right. When you turn on Interpolation in the Level module, you get audio rate interpolation. (Checking Interpolate in the Oscillator interpolates the waveform; it doesn't interpolate the amplitude *envelope*)

posted 01 August 2005 11:47            

quote:

---

Keyboard events are associated with one end of the pen; continuous controllers are associated with the other end. (By default, the eraser end is the Keyboard). You cannot have key events and continuous controllers on the same end of the pen. The only exception to this are the "tilt" controllers which work on both ends of the pen.

---

Dear SSC,

Here I am again, with this subject in mind.
I have been thinking about this a lot, and I would like to have user's opinion to know if what I try to achieve is useless or not. Maybe there is something I have not understood yet, so please notify me!

I think having the ability to map chosen continuous controllers on the same end as the Keyboard events end of the pen is very important, here is why: If I design a sound with the pen, I could try the same sound very easily on the keyboard and vice versa. On the other hand I could design a sound for the keyboard and quicly try it and enhance it with another parameters controller by the pen.

The problem is that if I chose to control a continuous controller event with an non-parametered gesture, then I lose the ability to control the keyboard events of that sound. Hence, I would have to remap every parameters from !KeyPitch and !KeyVelocity to "(!PenX * 24 + 60) nn" and !PenZ. This is be very tedious to do this and doesn't encourage to use the pen or keyboard.

In my personal opinion, I think that (if this is possible) pen and keyboard should be made "compatible", i.e. Keyboards events and Continuous Controller events should be treated the same way and both exposed in the Pen Window, allowing for quick accessibility and parameterization.

Please tell me if that makes sense.

posted 01 August 2005 14:28            

One solution might be to use two Graphire pads (and two pens) on the same USB bus. Use one of them as the keyboard and the other one for continuous controls. That way, your key events and continuous controllers would be independent of one another.

Another solution is to flip the pen around when you want to switch between key events and continuous controllers.

posted 01 August 2005 19:27            

quote:

---

Originally posted by SSC:
The problem with what you propose is that pairs of parameters would be linked to each other. For example, changes to !KeyPitch would *always* be paralleled by changes to !PenX. They would no longer be independent parameters.

---

Actually, I think there is a solution where one can choose where on the pen to put keyboard event, making them independant parameters. If keyboard events (if they exist in the defined sound) could be bound to any location on the pen (by using the pen window), then one for example could also choose to bind a continuous control to !PenY or any appropriate "gesture".

On the other side, the "tilt" solution seems good, although !PenY still can't be bound while using keyboard events on the same side of the pen, which makes it a "unused" parameter.