Re: Getting greatest decimal accuracy out of G_PI

[David Nečas (Yeti) <yeti physics muni cz>]

On Fri, Feb 02, 2007 at 05:50:47PM -0500, zentara wrote:
> 
> I've tried printing G_PI with the %Lf.20 format and I'm still stuck with 
> 15 decimal places of accuracy.

The definition of G_PI is IMO a bit misguided.  It has 50
decimal digits, but I cannot imagine how this excessive
precision could be used (except putting it into a string and
parsing that) because long double literals need an `l'
suffix:

#define MY_LONG_PI 3.1415926535897932384626433832795028841971693993751l
                                                                      ^here

Yeti


--
Whatever.