[Dev-C++] RE: [Mingw-users] GMINGW GCC 3.1 another defect?

[Ioannis V. <no...@ho...>]

> -----Original Message-----
> From: Mikael Aronsson [mailto:mik...@te...] 
> Sent: Wednesday, June 12, 2002 10:35 AM
> To: Ioannis Vranos; 'mingw-users'
> Subject: Re: [Mingw-users] GMINGW GCC 3.1 another defect?
> 
> 
> Hi !
> 
> You might wanna call it a defect, but there is not much that 
> can be done
> about it, the MS runtime does not support long doubles, it's 
> as simple as
> that.


Yes but VC++ 6 works fine with long double without producing erroneus
results (if i recall correctly from some years ago, then with SP3). Well
if MS doesn't provide long double, it is still a defect since it
produces erroneous results. long double could be equivallent to double.


> 
> What you could do, would be to force gcc to make all "long 
> double" normal
> "double" instead, and as far as I know that is fine with the 
> ansi standard
> also, but then you can't use long doubles at all.



It is fine with the standard and yes you could still use them. Intel C++
6 has double==long double and is considered one of the best compilers.



Sorry for insisting, but there can be no "excuse", the production of
false results is a defect...



Ioannis

* Ioannis Vranos
* Programming pages: http://www.noicys.cjb.net
* Alternative URL: http://run.to/noicys