Re: Information
On 2013-09-06 10:30:50 +0000, Corliss, George wrote:
> In this discussion, I agree with Bill Walster. In my opinions (other
> have and will differ), we have seen no application for which an
> exact result is essential AND for which people will pay money.
I also agree.
Moreover, some form of exact arithmetic may be useful, but should
be separate from a standard on interval arithmetic and should not
be limited to the dot product. In particular, it would be strange
to require an exact dot product, but not even an exact product.
And exact polynomial evaluation could be useful too. And what about
generic arbitrary precision, which could be more efficient (and
still provide exact results in particular cases)? So, I think that
all this should have its own standard. Whether it is implemented
or not (either in hardware or in software, or a combination) would
not affect the status of interval arithmetic, so that this would
be better for us than requiring operations that are not related to
interval arithmetic.
--
Vincent Lefèvre <vincent@xxxxxxxxxx> - Web: <http://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)