Re: [gpsd-dev] ✘CEP(50) 0.285 metersMime-Version: 1.0

[Gary E. Miller]

Yo Hal!

On Fri, 22 Jul 2016 17:10:58 -0700
Hal Murray <address@hidden> wrote:

> address@hidden said:
> > I'd rather do "long double" if I can.  Then no worry about improper
> > scaling. And I don't want to rewrite the Python gps.py library in
> > fixed point. Execution speed is not an issue.  
> 
> I'm not a language geek.
> 
> Wikipedia says:
> 
>   In C and related programming languages, long double refers
>   to a floating-point data type that is often more precise than
>   double-precision.
> 
> Notice the "often" in there.

As previously noted, in C11, it is 'recommended by the standard'.
gcc allows me control over what a long double is.  I could solve
the problem in C.

But that is way off in the weeds.  The immediate problem is solely
in Python.

> Maybe useful info here:
>   http://docs.scipy.org/doc/numpy-dev/user/basics.types.html
> 
> There is a float64 but no float128

I hate to pull in another dependency.  Unless it directly solves
the problem.

Since the root problem is computing small distances from very close
(lat,lon), do you know a simple libary to do just that?

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        address@hidden  Tel:+1 541 382 8588

pgpWJ7E_Du7x0.pgp
Description: OpenPGP digital signature