Gravity Z Component Difference about i2cdevlib

[jrowberg]

Hi choolake! I was unaware of the method InvenSense used in MotionApps to calculate the Z component of gravity. I used code that I came across as part of a different software IMU algorithm because it's what I had, and it worked. However, if the other method also works, there is no reason not to use it if you prefer. I'm not sure if one method has a performance gain over the other.

[drastef]

Hello all,

They are actually same formula. The formula given by litterature is Jeff's :
w^2-x^2-y^2+z^2. For a normalized quaternion, we have w^2+x^2+y^2+z^2 = 1.
The z gravity can then be written
(w^2^-x^2-y^2+z^2) -1 + (w^2+x^2+y^2+z^2) = 2w^2+2z^2 - 1, which is the formula used by Invensense (scaled by 2^30).

[jrowberg]

Thanks, @drastef! That's great to know. I will mark this issue as closed and update the code, since it looks to be slightly less computationally intensive.

[choolake]

I got your point and i am very grateful to jrowberg and drastef