Sender:	nate AT cartsys DOT com
Message-ID:	<376839E6.508A57AD@cartsys.com>
Date:	Wed, 16 Jun 1999 16:57:26 -0700
From:	Nate Eldredge <nate AT cartsys DOT com>
X-Mailer:	Mozilla 4.08 [en] (X11; I; Linux 2.2.10 i586)
MIME-Version:	1.0
To:	djgpp AT delorie DOT com
Subject:	Re: Floating point..... I think....
References:	<Pine DOT SUN DOT 3 DOT 91 DOT 990615094647 DOT 2646F-100000 AT is> <01beb7ba$28c43560$fc4484ce AT sub>
Reply-To:	djgpp AT delorie DOT com

0/0 wrote:
> 
> Eli Zaretskii <eliz AT is DOT elta DOT co DOT il> wrote in article
> <Pine DOT SUN DOT 3 DOT 91 DOT 990615094647 DOT 2646F-100000 AT is>...
> >
> > On 14 Jun 1999, 0/0 wrote:
> >
> > > I'm playing around with FFTs, and I notice my DFT and FFT functions
> become
> > > less acurate when I compile for DJGPP then they do when I compile with
> > > watcom and Visual C++...
> >
> > Did you link with "-lm" in the end of the command line?
> > Did you #include <math.h>?
> >
> 
> Yes I did "#include <math.h> and No I didn't use "-lm" but when I did it
> now, it had no effect.
> 
> I guess there could be something else going on, I'm going to try diffrent
> optimization functions just to be on the safe side, thank you, though.....
> BTW here are the comand lines I have tried....
> 
> gcc ft.cpp image.cpp main.cpp waveform.cpp -o ft.exe
> gcc -O3 ft.cpp image.cpp main.cpp waveform.cpp -o ft.exe
> gcc -O3 ft.cpp image.cpp main.cpp waveform.cpp -o ft.exe -lm
> 
> I'm going to see if it may be relating to something, with processor
> specific settings....

Probably not.

You might try to reduce it to a specific example where GCC and other
compilers give different results (for instance, write numbers to a file
and compare them), and then post it with your source and someone may be
able to track down the reason for the difference.
-- 

Nate Eldredge
nate AT cartsys DOT com

- Raw text -