pitch modulation in pattern

[Taron]

This is not an FR, rather a suggestion for an improvement:

- Pitch bend modulation in the modulation lane of a note pattern has a center line. It's almost impossible to manually hit zero there, because the mouse position to determine the value seems relative to the bottom or top border (which ever is 0% in your routine). Can you make the mouse position for data entry of +/- parameters like pitch relative to that center line, which is always at ZERO and can then be manually hit easily!? That would help a great deal and has always bothered me.

Just to reiterate: consider the actual pixel location of the center line as 0. Like (mouse ypos - centerline ypos)/(height of modulation lane * 0.5) = value.

Great many headaches will vanish at an instant!

[pljones]

Have you tried the event list editor when you need exactness?

[Taron]

Yes, but that's all hell when there's already more data in a pattern. I do this every now and then, when I'm desperate, but avoid it with a passion!

[MuTools]

What about a vertical grid there, cfr the vertical grid we have in the breakpoint envelopes?

[Taron]

Grids are good, but it's really about mouse input simplicity, not being at the mercy of pixel quantization around vital values such as 0.
The good thing about an improvement there would be that it was good under all circumstances, including curve design. It's rather a surgical improvement, which may be fairly painless to consider.

[MuTools]

Grids are good, but it's really about mouse input simplicity, not being at the mercy of pixel quantization around vital values such as 0.

That sounds too theoretic/complex for me. Can you pls give a simple concrete example that explains why a grid is not good enough. Thx.

The good thing about an improvement there would be that it was good under all circumstances, including curve design. It's rather a surgical improvement, which may be fairly painless to consider.

I think that if you want bezier/breakpoint functionality in the Event Graph Editor then i should rework the system so that pitchbend/controller/parameter curves are always done using breakpoint envelopes. Already possible now, but as an alternative to sequencial event curves. Implementing a single uniform system would also mean that when you do a live pitchbend/controller recording, the recorded events have to be translated into a breakpoint as good as possible: not too many points, not too few points. And there are more aspects. Conclusion: I'm open to this, think it would be an improvement if fully worked out, but it's not a quick 123 done.

[Taron]

I know, that's why I'm just asking for a very simple mouse coordinate interpretation for values.
I'm assuming that you currently do the following:
Mouse_y = mouse y position
Lane_y = modulation lane y position
Lane_height = height of modulation lane

(Mouse_y - Lane_y) / Lane_height = Value as percentage between 0 to 100% then multiplied by what ever the parameter range asks for.

I'm not sure which way around you have your y coordinate system, so the y position of the modulation lane I'm talking about is either on the bottom or the top. If it was upside down then you'd invert the value...yadda yadda, you know what I mean, I'm sure.

I propose to make this value formula relative to the center of the modulation lane, if the parameter is bipolar like pitch bending, for example.
(Mouse_y - (Lane_y+0.5*Lane_height))/(0.5*Lane_height) = Value as percentage between -100% to 100%....

That would virtually require nothing but a check for polarity (unipolar or bipolar) and then the use of the alternative mouse to value interpretation.
This same logic could remain for ever not just for straight value input but also including when it comes to positioning curve points, meaning that it would never be a waste!

[MuTools]

I know, that's why I'm just asking for a very simple mouse coordinate interpretation for values.

That's snap to grid. No?

I propose to make this value formula relative to the center of the modulation lane, if the parameter is bipolar like pitch bending, for example.
(Mouse_y - (Lane_y+0.5*Lane_height))/(0.5*Lane_height) = Value as percentage between -100% to 100%....

Even without looking at the actual code whether it's unipolar/bipolar (yes i understand what u mean), it would still mean that there is only a single pixel on the zero, which is hard to get. That's why i think a grid is a better solution. No?

[Taron]

Snap to grid is great for curve point modifications. I'd be the first to salute with elation the introduction of curves for everything, I mean...I'd go crazy for joy, especially if they were Bezier!
I just thought of a quick fix for the currently existing value input down there. A grid snap would not work, because it would make it even less pleasant to "draw values" freely.
Hitting one pixel is no trouble at all, especially when you use such a formula, by the way. The moment you see 0.0 show up, you have enough hope to nudge right on it with joy!

[MuTools]

Ok i see your point. Thx for elaborating.

[Reincke]

What about a vertical grid there, cfr the vertical grid we have in the breakpoint envelopes?

We're talking about the sequence's event graph editor, right? A configurable vertical grid there would be a great step ahead! Pitch bends will get easy.
JR

[Taron]

HAHAAAA! Pandora's box! Told ya'!
I know, right, imagine you could make a grid based on note frequencies, hehehe....could get pretty unbelievable, right!
Even without a grid snap this would be brilliant orientation.

[AndreasD]

For a not direct but workable solution there is this:

Pitch Selector.jpg

Then edit to "0" in the list editor.