few basic questions

I’m working on a html5/webgl 3d virtual tour app(specially for windows OS). right now my app works but it has some limitations. it has to run on goolge chrome and of course some vga cards doesn’t support webgl. but i need my app works on any system as much as possible.
So i have these few questions before i start to re-implement my app in panda3d:
1- Does panda3d have such a blacklist vga cards like webgl? or it works on almost every vga cards?
2 - My model designed in blender and it has textures, baked lightmaps and some transparent textures. so how much of it would be exported into panda3d egg file?
3- How hard could it be to implement collision detection in a room with walls and objects inside it? i mean i don’t want to camera passes through the walls or generally everything in my model.
4- I need an special fps camera control which should has a mouse cursor and camera rotation should be done by right-click drag. is it possible?
5- When user clicks on an object inside my model i want to show some info form on the screen. i know there is a gui module for 2d design, but since i already have bunch of html forms and panda3d has a browser plugin, i’m thinking of using panda3d plugin inside my html5 app. but how can i distribute panda3d plugin? it has to be run by installer or is it enough to put dll/ocx beside my app root and create activex instance by javascript?

Sorry if i asked too much. that’s because i have short time to convert my webgl app into something better.
I know python and i love it.
Thanks in advance.

I have answers to two of those questions, I believe:

This might depend on just how complex your scene is, whether you want advanced physics, and whether you have any special requirements.

However, for a simple scene and fairly straightforward collision this shouldn’t be too difficult.

Oh yes, very much so, I do believe. In short (and simplifying a little), Panda should allow you to detect the right-mouse button being pressed and released, allowing you to set a “dragging” state. With that, the current mouse-position, and the current delta-time (the latter two of which Panda should also provide) you should be able to determine a rotation to apply to the camera. Depending on how you want this to work, you may also want to reset the position of the mouse-cursor to the centre of the screen.

Great, that’s what i expected.
But I’m more concern about range of video cards support.

It should work on any card supporting OpenGL or Direct3D 9. There is no blacklist, but if you find a card on which Panda doesn’t run, please let us know, and we’ll do what we can to fix it.

It depends on how the textures are assigned to the material. These things you name are supported in principle, but it might take some tweaking of the settings in order for the export to work correctly.

A sample program demonstrating these things:


People will need to install the Panda3D runtime, which is a small installation (a few megabytes). You can redirect people to panda3d.org/get to install it. We also have some scripts in the repository that can be used to generate a .cab or .xpi for installation as an add-on in IE or FireFox, but I haven’t personally tried that.

Ok, i’ll go for panda3d! actually i already started to compile it.
My target graphic card which seems doesn’t support webgl is intel 82945g chipset family. unfortunately i haven’t access to that PC right now to test panda3d on it. but i will try.
Thanks for reply.

Hmm, that seems to have an Intel GMA 950 for integrated graphics. It supports Direct3D 9 and OpenGL 1.4.

I found this thread referencing it:

That suggests that Panda works on the GMA 950. Though it’d be a good idea to test it anyway. (If it doesn’t work for some reason, it might be necessary to use config switches to disable some more advanced OpenGL features. In the worst case, you could switch Panda to use the Direct3D renderer back-end.)