Re: Verdix VADS ADA or GNU ADA on an SGI...

[merlin@annwfn.com (Fred McCall)]

In <1994Sep22.143838.388@mksol.dseg.ti.com> connor@mksol.dseg.ti.com martin connor writes:

>We are in the process of selecting an ADA compiler for an SGI workstation.
>The engineers want VERIX VADS, the accountants want GNU ADA (based on cost).
>Has anyone used either of these on an SGI workstation?  Does GNU ADA have
>tools, or support?  In the long run would the VERDIX ADA be a better choice?

In DSEG you don't really have the option of using GNAT.  If you're
working on something where you have to use Ada (which is, I presume, why
you're using it if your beancounters are that concerned about compiler
costs) you're going to have to use a validated compiler.

I also don't recall hearing that it had been ported to SGI IRIX (last I
knew, GCC wouldn't build correctly, but that was a year ago so things
have probably changed somewhat in that regard).

If you're writing MCCR software (or probably anything that falls into
deliverable software, for that matter), you have to use a validated
compiler suite.  You also won't be able to get any support from SGI for
any problems you may have (and, using an unsupported compiler, you're
going to run into lots of problems where your support for GNAT will say
it's an IRIX bug and SGI will tell you that you're using an unsupported
compiler so it's your problem).

[Can you say 'captive market'?  Even given access to free tools, it
still exists as long as we have to deal with the Ada Mandate.]

--
"Insisting on perfect safety is for people who don't have the balls to live
 in the real world."   -- Mary Shafer, NASA Ames Dryden
---------------------------------------------------------------------------
merlin@annwfn.com -- I don't speak for others and they don't speak for me.