kicad-developers team mailing list archive

[Tomasz Wlostowski <tomasz.wlostowski@xxxxxxx>]

On 25.01.2015 02:50, Miguel Ángel Ajo wrote:
> 
> I was just testing that.
> 
>   Python floats (in all architectures) are double, so the 52 bit fraction (11 bit exp) is always
> going to hold more information that the 32bits wxPoint /wxSize can hold.

Hi Miguel,

Sorry, I'm a C/C++ guy, so float defaults for me to 32 bits ;) Doubles
look ok, although the decimal part will be anyway truncated on the way
from Python to C++ (if we decide to keep the nm as default unit)

Cheers,
T.