Yes. I think that using the dither matrix based on the R_2 sequence should be competitive with any of the the dither masks mentioned on that link.

Thus in situations like video, and/or ray tracing where computing speed is of the essence, maybe a better mask such as the R2 dither mask would offer an improvement on existing methods.

However, in terms of end-quality I am not aware of any dither masks that are anywhere near as good as the state-of-the-art error diffusion algorithms. Thus, for a small number of images such as our favourite animated gifs, I would still be recommending an error-diffusion method. :)