Mail Archives: djgpp/2002/02/18/08:08:38
cesar tejeda <cesar_tejeda_her AT yahoo DOT es> wrote:
> TurboC needs a lot less compile time for the same file(10 times less
> approx.) , and it is also a FASSSTER environment when you compare it
> to RHIDE. It also uses a lot less memory.
That's an old and well-known fact. But it also has it's backside:
Turbo-C's compiled programs themselves are a *lot* slower in many
situation that what GCC generates. Especially if you happen to just
need raw CPU power and lots of memory more than direct access to
hardware registers.
For short: the "Turbo" in TurboC only applies to compilation speed,
but almost anti-applies to executable speed.
> I suppouse that efficiency is not one of the targets for gcc
> compiler.
It is. But it's the efficiency of the compiled program, rather than
the compilation process itself. That, and the traditions from the
Unix worlds it sticks to a lot more strictly than TurboC.
> My 386 has only 2MB memory, perhaps it is the worst environment
> where DJGPP has runned in. ;-)
Definitely not. It may be about the lowest it *still* is actively run
in. But in earlier days, 386's with about that amount of memory were
quite commonplace, even among the DJGPP core team.
Esp. those 2MB are a serious problem as it comes to running DJGPP.
4MB and a clever cache+ramdrive strategy could make a world of
difference, already. You're currently bottlenecked by the speed of
your *harddrive* more than your CPU or GCC itself.
--
Hans-Bernhard Broeker (broeker AT physik DOT rwth-aachen DOT de)
Even if all the snow were burnt, ashes would remain.
- Raw text -