Quaternion vector rotation optimisation

[legends2k]

The current code

template <typename T, precision P>
    GLM_FUNC_QUALIFIER detail::tvec3<T, P> operator*
    (
        detail::tquat<T, P> const & q,
        detail::tvec3<T, P> const & v
    )
    {
        T Two(2);
        detail::tvec3<T, P> uv, uuv;
        detail::tvec3<T, P> QuatVector(q.x, q.y, q.z);
        uv = glm::cross(QuatVector, v);
        uuv = glm::cross(QuatVector, uv);
        uv *= (Two * q.w);
        uuv *= Two;
        return v + uv + uuv;
    }

does two multiplications of 2.0 which is avoided in this implementation

    const detail::tvec3<T, P> axis(q.x, q.y, q.z);
    return v + T(2) * glm::cross(axis, glm::cross(axis, v) + q.w * v);

This implementation is based on these two sites https://code.google.com/p/kri/wiki/Quaternions and http:https://molecularmusings.wordpress.com/2013/05/24/a-faster-quaternion-vector-multiplication/.

[Groovounet]

This issue is fixed in GLM 0.9.5 branch for GLM 0.9.5.4 release.

Thanks for contributing,
Christophe