Mouse movement curve radis.

[cmwb2000]

Hi

I am trying to test to see if there is a link between the curve of a mouses movement and the length from the ball to the wrist (pivet point). I am recording a series of points as the mouse moves approxamatly every 20ms. From this I need to try calculate a circle of best fit from idealy 2 points or 3. I was thinking start, middle and end positions. The calculations I have found online are somewhat over my level of knowlege.

So far the alternative that I have come up with is to measure the differance in degrees of movement between readings, divide this by the length to get length per degree and multiply this by 360 to get the circumfrance. However I know this to be the wrong way of going about this.

Any ideas and help would be greatfully recived.

Chris

[JosAH]

Doesn't the wrist move itself, i.e. is that pivot point on the wrist a stationary point?

kind regards,

Jos

[cmwb2000]

I am trying to develop a series of tests to try and evaluate a uses identiy through the mouse, I noticed that I never move my wrist. I only move the mouse with my hand from piviting around my wrist. Hence I want to see if the radius of a mouse movements curve is directly related to the actual distance from the mouse ball or led to the wrist.

Chris

[JosAH]

@cmwb2000
Ah, ok, here's a bit of simple math then: assume you sample three points P1, P2 and P3. Given points P1 and P2 there's a midpoint M12; the same goes with points P2 and P3: there's a midpoint M23. The lines perpendicular to the lines P1 and P2 (P2 and P3) running through the midpoints cross each other at the centre of a circle and points P1, P2 and P3 are on the circumpherence of that circle, given that the wrist doesn't move. Does this help you a bit?

kind regards,

Jos