Handling Multitouch

First off, I just starting using SIO2 and so far it looks to be amazing. Great work.

I am trying to figure out how to handle multi-touch and cant seem to find a method for this. Looking at the EAGLView touchesBegan, touchesMoved and touchesEnded methods and the relevant sio2 events it doesn't appear that the touch count gets passed along to the event. I may be way off on this as I have only been playing with sio2 for a few hours. How do you handle multi-touch events with sio2?

Thanks

Ho! I see what you mean... No SIO2 doesn't support that, this is application specific, just implement your own routine inside the tap function and use a global variables to hold the # of touch currently on screen... Lemme know if you have any problems, by the way is it something that you guys would like to see in the next revision or?

Hope this helps,

Cheers,

sio2interactive wrote:... by the way is it something that you guys would like to see in the next revision or?

sure!!! ^^ (and if we can suggest... some animation system would be the coolest thing, ah! and don't forget normal maps! )


btw, I'm still working on the 3dsmax exporter, when is going to be released the next revision? so I can know if I can have it done for then...

sure!!!

>> Once again Im not sure that it should be implemented globally, take a look in the case of the widgets, which touch activate the widget 1, 2 or 3? that doesn't make sense... I'll try to think about a generic approach that would satisfy everybody (as always) but I need suggestions in order to handle every possible cases.

^^ (and if we can suggest... some animation system would be the coolest thing,

>> Of course I can hardcode right now MD2/3 or MD5 support... but is it really the best solution? What about full IPO integration? for camera, lamp, object and armature + vertex group with NLA support & action selection? That would be the real deal... so to do all that and test and integrate it properly that takes time Unfortunately animations are not on top of my list... Cuz if I release animations I want to do it right and provide FULL integration on everything in the scene.

ah! and don't forget normal maps! )

>> You can already do it, but most likely SIO2 will never have it embedded, there's no shaders (and that would be SO easy to do with shaders) on the devices... so imagine a sphere with 4 lights around it, the dot3 combiners need an TBN matrix for every lamp and assigned on the vertex color... so what happen when 4 lamps are around, every triangles have to be determined on which lamp is the nearest and how much this triangle is affected + the sphere have to be rendered 4 times excluding triangles that are visible / not visible from the FOV of every lamps (4 rendering pass)... that's way too much for the device... There's no generic approach that would satisfy everybody and would not affect the overall performance. So I believe that is more application specific... and should be implemented manually. If you guys want I can always provide an example how to do it... it would never be supported natively since there's no generic approach or method that can be applied on a large scale.