Re: Current Usses for Fortran ??
From: Richard Maine (nospam_at_see.signature)
Date: 08/25/04
- Next message: Keith Refson - real email address in signature: "Re: Question about precision"
- Previous message: Herman D. Knoble: "Re: Question about precision"
- In reply to: Danguard: "Re: Current Usses for Fortran ??"
- Next in thread: Pawel: "Re: Current Usses for Fortran ??"
- Reply: Pawel: "Re: Current Usses for Fortran ??"
- Reply: Danguard: "Re: Current Usses for Fortran ??"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 25 Aug 2004 08:33:27 -0700
Danguard <danguard_robot@hotmail.com> writes:
> So, my question is: why you use F77 and not a more modern Fortran??
Mostly I don't use f77, but use the more modern versions instead.
I gave up on f77 nearly 15 years ago, even a little before the first
f90 compiler was out. Although I knew f77 well and had extensive
experience in it, I concluded that it was just too awkward for the
new projects I was doing then.
But as to why people would use f77, the main reasons are:
1. They are working with old codes written in f77. This is a big one;
it is a reason that still sometimes applies to me.
2. The availability of free f77 compilers is wider than that of f90
ones. Note the word "free". There are plenty of f90 compilers
for sale. In fact, it is getting harder to buy new compilers
restricted to f77 than ones that also do f90 or better. (All
f90 comilers can also do f77). This reason is gradually going
away. You can now get free f90 compilers for Windows and Linux.
More are on the way. (The free ones do have limitations, but
if you don't fit in those limitations, there really shouldn't
be much an issue with the cost of most of the commercial
compilers).
3. Some people are just used to f77 and don't want to change or
feel the need to. I'd label this as probably the single largest
real reason. Sometimes I'm sure this is the underlying reason
when other factors are cited.
Please note that I do not necessarily imply anything negative about
people with this reason. For some people, in some situations,
this may be the best choice. There are certainly choices in life
where I have decided that I don't want to change an old practice
or particularly feel the need to. Fortran just doesn't happen
to be one of the places where I made that kind of choice.
4. Some people are forced to use f77 by outside constraints (management,
customers, whatever). Those outside constrants may in turn arise
from any of the previous 3 causes. I had this happen to me on one
relatively recent project that was subsequently cancelled (X-33).
I'm convinced that in that case, it was purely a prejudice of a
particular manager; if I had told him that the tool I preferred to
use was a kind of auto-coder for C, then it would have been acceptable
as long as I didn't mention the term Fortran 90. Other C auto-coders
were used in parts of the project, as was f77, but f90 was verbotten.
Oh, and arbitrary f77 extensions were ok, as long as the word f90
wasn't mentioned. The f90 compiler I was using could have been
accurately described as a C autocoder, though it isn't the way I'd
normally describe it.
-- Richard Maine email: my last name at domain domain: summertriangle dot net
- Next message: Keith Refson - real email address in signature: "Re: Question about precision"
- Previous message: Herman D. Knoble: "Re: Question about precision"
- In reply to: Danguard: "Re: Current Usses for Fortran ??"
- Next in thread: Pawel: "Re: Current Usses for Fortran ??"
- Reply: Pawel: "Re: Current Usses for Fortran ??"
- Reply: Danguard: "Re: Current Usses for Fortran ??"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|