On Mon, Nov 29, 2010 at 9:11 PM, Scott McMurray <me22.ca+boost@gmail.com> wrote:
On Mon, Nov 29, 2010 at 07:24, Hal Finkel <half@halssoftware.com> wrote:
On Mon, 2010-11-29 at 11:07 +0100, Mathias Gaunard wrote:
It would be nice to know what minimum value is necessary to provide correctly rounded results for IEEE754 floats and doubles.
This depends on the number: For a rational, you need only two integers ;) For a general real number, it depends on the rate of convergence.
Wouldn't it be bounded by the number of terms needed for a sufficiently-precise Phi, since it's the slowest-converging CF?
For some inexplicable reason, the very fact that there's someone on this list who knows enough about math to ask that question makes me really, really happy. -- Dave Abrahams BoostPro Computing http://www.boostpro.com