[Numpy-discussion] longdouble (float96) literals

[josef.pktd at gmail.com]

On Wed, Aug 18, 2010 at 10:58 AM, Charles R Harris
<charlesr.harris at gmail.com> wrote:
>
>
> On Wed, Aug 18, 2010 at 8:54 AM, Matthieu Brucher
> <matthieu.brucher at gmail.com> wrote:
>>
>> I don't think there is longdouble on Windows, is there?

>>> np.longdouble
<type 'numpy.float96'>

>>> np.longdouble(5).itemsize
12
>>> np.double(5).itemsize
8

>>
>
> The MSVC compilers don't support extended precision, or rather, long doubles
> are the same as doubles. I don't know what other compilers on windows do.

I have the mingw version, gcc3.5 or so

So to me it looks like float96 is there, but not used in calculations.
(I never tried to use it for anything before.)

Josef

>
> <snip>
>
> Chuck
>
>
> _______________________________________________
> NumPy-Discussion mailing list
> NumPy-Discussion at scipy.org
> http://mail.scipy.org/mailman/listinfo/numpy-discussion
>
>