>
> Isn't x guaranteed to be 0.0F here?
Yes, it will 0.0f but its not used in any computation anywhere. So new
'else'-block should be treated as (1,0,0) on input - its optimized
computation of rotation only by x.
On Tue, Sep 18, 2018 at 5:56 PM, Gustaw Smolarczyk
wrote:
> wt., 18 wrz 2018 o 15:
wt., 18 wrz 2018 o 15:59 Sergii Romantsov
napisał(a):
>
> Specification doesn't define behaviour for rotation of 0-vector.
> Windows and Nvidia drivers have a workaround for that.
> For compatibility proposed that for 0-vector a rotation will be
> done around x-axis.
>
> -v2: logic moved to _math_
Uploaded new version.
Seems that waiting for any response on
https://github.com/KhronosGroup/OpenGL-API/issues/41 may take much more
time
On Tue, Sep 18, 2018 at 4:57 PM, Sergii Romantsov <
sergii.romant...@gmail.com> wrote:
> Specification doesn't define behaviour for rotation of 0-vector.
>
Specification doesn't define behaviour for rotation of 0-vector.
Windows and Nvidia drivers have a workaround for that.
For compatibility proposed that for 0-vector a rotation will be
done around x-axis.
-v2: logic moved to _math_matrix_rotate
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?i
