How to do frame rate independent mouse?

[MasterGod]

For example I want the mouse movement speed to be the same in 640x480 resolution and in 1600x1200.
I guess a solution would be to check with percentages but I don't have a clue on how to do that?
Or do you have a better solution?

Thanks.

[Ion Dune]

Do you mean that it takes the same physical mouse movement to get the mouse to move from one side of the screen to the other, or for camera rotation like an FPS camera? For the latter, getRelativePosition, but I suspect you mean the former.

The simplest way I can think of doing it is to use trig to find the direction of movement, then recalculate how far it would have gone using a timer and a relative position.

[MasterGod]

The simplest way I can think of doing it is to use trig to find the direction of movement, then recalculate how far it would have gone using a timer and a relative position.

Sorry but I still don't get how

[danieLs0xFF]

i think separate thread for mouse handling would do the trick

[MasterGod]

i think separate thread for mouse handling would do the trick

Don't you think that's too complicated for such a simple task??

[rogerborg]

i think separate thread for mouse handling would do the trick

Don't you think [...]

I doubt it.

Aaaanyway, to repeat the question: do you mean the movement of the cursor across the screen, or the effect of that movement on (e.g.) a FPS camera? Can you please be specific about your requirements.

[MasterGod]

Aaaanyway, to repeat the question: do you mean the movement of the cursor across the screen, or the effect of that movement on (e.g.) a FPS camera? Can you please be specific about your requirements.

Ok, I'll try to describe it better.
Let's say I have a sphere mesh.
When I raise the mouse up it rotates on -X and down on +X.
The mouse cursor is locked to the center of the screen.
Now, if I move the mouse up fast it moves lets say 10 unites. if I move the mouse slow, the same distance on my desk, it moves 75 unites. (10 and 75 for example).

P.S
I'm starting to think now the problem might not be with making the mouse "time-based" but with something else..

[robertgarand]

Hi,
I m using the mouse to direct a cannon on the screen,
I use factor to multiply actual mousemov. it is currently
hard coded ( the 70's), but one culd just implement different
factors base on current screen resolution.
hope it helps.

Code: Select all

//.5 is center of screen in absolute mouse pos
const f32 anglecanon = 0.5;


	video::IVideoDriver* driver = Device->getVideoDriver();
	core::vector3df canonrot;
	f32 offsetx, offsety;
	if (!Device->isWindowActive()) {return;}
	gui::ICursorControl* cursorcontrol = Device->getCursorControl();
	core::position2d<f32> cursorpos = cursorcontrol->getRelativePosition();
	offsetx = (cursorpos.X-anglecanon)*70;
	offsety = (cursorpos.Y-anglecanon)*70;
	core::vector3df canon_rot= (canon->getRotation());
//X & Y is inverted on screen resolution versus gui mousePOS
	canonrot.Y =offsetx;
	canonrot.X =offsety;
	canonrot.Z=0;
	canon->setRotation(canonrot);

Hope it helps.
Robert

[Ion Dune]

Oh, I see. So the same distance on the desk produces different results based on speed? I didn't even know it worked that way.
If you moved the mouse one centimeter physically, and that equaled 10 pixels on the screen, are you saying that if you move the mouse another centimeter but faster it will move less than or more than 10 pixels. If so, I don't think its something that can be changed in Irrlicht. It sounds like a problem in the operating system or the mouse.

[MasterGod]

Oh, I see. So the same distance on the desk produces different results based on speed? I didn't even know it worked that way.
If you moved the mouse one centimeter physically, and that equaled 10 pixels on the screen, are you saying that if you move the mouse another centimeter but faster it will move less than or more than 10 pixels. If so, I don't think its something that can be changed in Irrlicht. It sounds like a problem in the operating system or the mouse.

No, it doesn't happen outside Irrlicht, or at least my app (didn't tested it on the examples yet).
On windows/GTA SA 10 cm equals the same distance no matter what speed I move the mouse.

I'll check the behavior on the examples tomorrow and report back.

@robertgarand: I'll check it tomorrow, thanks.

[Ion Dune]

Ahh! I see now. I believe that I had a similar problem once, and that I was able to fix it by going from cursorcontrol->getPosition to getRelativePosition.

[MasterGod]

Ahh! I see now. I believe that I had a similar problem once, and that I was able to fix it by going from cursorcontrol->getPosition to getRelativePosition.

It didn't do the trick.

Maybe the following video will describe good enough my problem:
http://www.youtube.com/watch?v=WX2moCiCuEg

[Ion Dune]

Huh. Well, I am stumped. Perhaps you should compare your code to that of the FPS camera scene node. It should be the same sort of mechanism.

[MasterGod]

I tried but I can't figure which code inside the FPS camera take care of such behavior.
Does anyone else knows what causes it?

[rogerborg]

(Sorry, I missed your earlier reply)

Answer: dunno. How are you retrieving and interpreting the mouse position/movement?