Can SM record Mouse Movement

oddson · 31 Jul 2006

You could track the location of a string of mouse events and store them as float arrays (x and y and possibly mouse state changes) but the problem will be timing events. The proposed event mechanism scheme (previewed in SMX) may make this sort of thing feasible but for now you'd have to rely on some kind of counter recording triggers from a Ticker to give time indexes for the recorded events. This would not be very accurate and could easily bog down from CPU load.

If you want to limit recording to changes in cotrol values you could do this more easily but again the timing of events would be very approximate.

If this doesn't dissuade you from the idea I could try to knock someting up as a demonstration.

sequencer · 31 Jul 2006

Thanks for the reply Oddson. 🙂

I guess I would have to wait for the timing event system to be upgraded.

If you have a few precious moments to spare, could you possibly put something together so that I would have a bit of an idea how to go about it. At least then, when the timing system improves, I will hopefully have a running system of what I need to do and can just convert what I need to the new event system.

Thanks 🙂

oddson · 31 Jul 2006

I'll give it a go... do you need to track mouse events or just contol changes?

Like I said above, the latter would be much easier to deal with.

sequencer · 31 Jul 2006

Well, I might as well spill the beans about what I'm thinking of trying to do. 😉

I've made a XY Pad for assignable MIDI CC output, so I was kind of thinking it would be good if I could record the movement of the position cursor and have it 'play' back whilst altering another. Is that possible yet ?

I just want to try to make useful tools for my own use and eventually (once I get round to getting a website up and running) to release (Free) on the general public if I think they are good enough.

Anyhow, if you manage to get something together, I would be extremely grateful. 😀

oddson · 4 Aug 2006

OK... as I suspected it can be done currently with Ticker events but it is very inacurate and easily overwhelmed by graphic event slowdown.

But you can see roughly what I had in mind

xypadRecord.osm

If you want to see that it works I recommend recording very slow movements so the ticker events are not slowed down too much while recording.

sequencer · 4 Aug 2006

Thanks for this...

I was beginning to think that it might have stumped even you. I should have known better.

Thanks for the work in the WIKI. I go into the WIKI to view info about the components and as a result see your sterling efforts.

😀

oddson · 4 Aug 2006

sequencer wrote: ...I was beginning to think that it might have stumped even you...

😊 Even me? That's a great complement. Especially when I think back to my early efforts and how lame they were.

In reality the hardest part of this was getting an XY pad to work for the deomonstration of the recorder. I'm sure yours will be be much better.

The recorder is pretty simple stuff. It's just filling arrays with control values.

sequencer · 4 Aug 2006

Alas, I won't get the chance to look into this at all for a couple of weeks. Off on a big silver (White with red & blue actually) bird tomorrow, for a two week stint in N.J. in the US.

Once I've finished working over there and get back to over here, I'll have a look properly.

Thanks 🙂