[eigen] Re: Non-commutative algebra

[Peter <list@xxxxxxxxxxxxxxxxx>]

Dear Christoph,


Am 23.12.2016 um 11:15 schrieb Christoph Hertzberg:
> We do at several points. E.g.
>   s1 * A * s2 * B * s3
> is optimized to something equivalent to
>   (s1*s2*s3)*(A*B)

o.k. thanks, I'll better stay with my for loops.

> You may be interested in this feature-request:
> http://eigen.tuxfamily.org/bz/show_bug.cgi?id=560
That's actually a different story. I don't need quaternion for that,
my types are actually Grassmann variables combined with a floating point.

Best regards,
Peter


Best regards,
Peter