Rui Maciel wrote:
Victor Bazarov wrote:
>The "best" compiler is the
one you use *successfully*, i.e. to create programs *that work*.
But what about standards compliance? Shouldn't they count something
on the evaluation?
Only if it matters *to you*. I've heard some folk are still
successfully using TurboPascal and love it.
Independent (from *you*) evaluation tends to be less and less
valuable if the criteria it uses are farther and farther from *your*
criteria. If the compliance is one of your own requirements (for
whatever reason you think is important), you add it to the list of
those on which you make your decision. What you essentially need is
a table in which there would be *all* possible criteria listed for
*all* compilers from which you have to select. Then you need to
hand-pick the rows (columns) and then examine your [reduced] set of
compilers WRT the hand-picked criteria. You also need to apply
weights to the criteria, most likely. Then, after a while, you will
get *your* best. Asking for somebody else's "best" is pointless.
Am I muddying the waters well enough?
V
--
Please remove capital 'A's when replying by e-mail
I do not respond to top-posted replies, please don't ask