Mail Archives: djgpp/2000/06/02/19:52:21
On Fri, 2 Jun 2000, Eli Zaretskii wrote:
> > Date: Fri, 2 Jun 2000 21:46:30 +0600 (LKT)
> > From: Kalum Somaratna aka Grendel <kalum AT lintux DOT cx>
> >
> > Well in the end what really matters is not the portability but how good is
> > the code that the compiler produces for a specific architecture.
>
> That is true.
>
> > It is of no use if the vastly architecturally portable compiler generates
> > tolerable code for the x86 platform...as anyone would prefer a less
> > architecturely portable compiler which generates better code which is
> > specifically tailored for the x86 chipset..
> >
> > Which is why the majority of people still use Watcom/MS C++
> > extensively for coding for the x86 platform.....
>
> I'm not sure I understand what are you trying to say here, but if it's
> that GCC produces code that is inferior to Watcom, then I suggest to
> take a look at the compiler comparison page,
> http://www.geocities.com/SiliconValley/Vista/6552/compila.html,
> maintained by Salvador Eduardo Tropea. I think you will be surprised
> (pleasantly, I hope ;-).
Actually I have had a look at the page but IIRC SET benchmarked it usign
the 10.x series and in a earlier thread sometime back one poster suggested
that the 11.x Watcom compiler generates much better code than the 10.x
compilers...
Also there was some question about what were the switches used for the
watcom compiler during this test.....
> This thread was about memory consumption by the compiler, not the
> quality of the code it generates. These are two different and
> uncorellated issues.
Agreed.....But what I commented about was that ultimately it is not the
memory consumption but the quality of the code that is generated that
matters..
> > Could anyone please tell me how many x86 architecture based exes out of
> > the many that you come across are compiled using GCC...much less than the
> > ones that are compiled using Watcom/borland/M$C and other x86 specific
> > compilers...
>
> Since when is the consumer base a reliable evidence about the quality
> of the product? Should I remind you the Windows vs Linux case?
Sorry if i am missing something, but I thought that the consumer base is
*the* reliable evidence of the ultimate quality of the product int he
long run.
And please note that for the user the "quality" of a product does not mean
the technical quality, but how user friendly it is. A fact which is
overlooked by the linux advocates. Windoze is so popular because it is
much more user friendly than linux and all it's GUI's.....
They expect that the secretary would prefer a technically superior linux
over windoze and then they expect her to do her word processing in ..VI !
Just see where GNOME got it's feel...yes from windoze 95.
for example I would prefer A "inferiror" product with a GUI based debugger
than a "awesome?" one with a *!*!! command line based debugger...
Grendel
Hi, I'm a signature virus. plz set me as your signature and help me spread
:)
- Raw text -