Lahey Support
08-15-2003, 01:33 AM
Your comment about the wide implementation of G77 because it is free and
easy to obtain is very true. It was students that carried "c" and Pascal
and the vi editors from the classroom and lab to the workplace. If we have
a free, readily available Fortran following up to date standards, with
support libs and of course some windows support, students would take that
with them to the workplace.
----------
From: Lawson B. Wakefield
To: [address removed]
Subject: Re: [LF] Fortran style
Date: Tuesday, September 04, 2001 5:33AM
In article [address removed] Szondi Egon Janos
[address removed] writes
>Jos L Gmez Dans wrote on Aug. 31:
>
>> I think that the existance of a free compiler
>>(http://g95.sourceforge.net) would be a great aid as well. I kno plenty
>>of people using f77 because of g77 (apart from the seamless integration
>>with gcc).
>
>There are two problems using g77:
>- The most important developments of Fortran 90 are lacking (e.g.
>modules, ALLOCATE statement, pointers, optional arguments...).
>For details see: Lahey Fortran 90 Manual, App. B. As a consequence,
>using g77, one cannot process up-to-date programs got from scientific
>information exchange sources. And, of course, he/she will not be able
>writing up-to-date programs.
While this is fair comment about g77, I think the point that was being
made was that the availability of a free full Fortran 95 (i.e. g95) will
help a lot. The g95 project is still at a relatively early stage and
many people still don't appear to be aware of its existence. From what I
know of the effort required to develop commercial f90/f95 compilers, I
think the g95 developers have a bigger job ahead of them than they
realise, but I wish them well.
>- This compiler is distributed according to the GNU license. It means,
>everybody has the right to modify the source. Therefore, there is no
>guaranty on the quality of the compiler. As a consequence, this
>compiler must not be used solving problems requiring quality assurance.
>(On the other side, this condition pronounces the responsibility of
>the compiler vendors - the users have to pay for it.)
Agreed, though in practice it's actually fairly unusual to encounter a
'doctored' g77 of this sort. Quality-wise g77 is sufficiently good that
a large number of people do seem to be using it, with considerable
success. I must admit I've never found it to be up to the standard of
commercial offerings like Lahey F77L-EM/32 or Salford's FTN77, but we've
successfully used it to build substantial code on a variety of platforms
with little problem.
Ironically, I would say that g77 is actually part of the problem when it
comes to encouraging usage of modern versions of Fortran. Its free
status, wide availability and adequate quality mean that its in quite
wide usage. This is apparent from the amount of g77-related traffic on
the comp.lang.fortran newsgroup. Essentially, there is a sizeable user
base (particularly students) who will use just about any compiler if
it's free and easy to obtain, regardless of the long term cost of using
a limited and obsolete language standard.
Lawson Wakefield
! Interactive Software Services Ltd.
! The Fortran GUI specialists
! Winteracter home page : www.winteracter.com
----------------------------------------------------------
To unsubscribe, send to [address removed] the following
as the first and only line of the message body:
unsubscribe fortran
----------------------------------------------------------
easy to obtain is very true. It was students that carried "c" and Pascal
and the vi editors from the classroom and lab to the workplace. If we have
a free, readily available Fortran following up to date standards, with
support libs and of course some windows support, students would take that
with them to the workplace.
----------
From: Lawson B. Wakefield
To: [address removed]
Subject: Re: [LF] Fortran style
Date: Tuesday, September 04, 2001 5:33AM
In article [address removed] Szondi Egon Janos
[address removed] writes
>Jos L Gmez Dans wrote on Aug. 31:
>
>> I think that the existance of a free compiler
>>(http://g95.sourceforge.net) would be a great aid as well. I kno plenty
>>of people using f77 because of g77 (apart from the seamless integration
>>with gcc).
>
>There are two problems using g77:
>- The most important developments of Fortran 90 are lacking (e.g.
>modules, ALLOCATE statement, pointers, optional arguments...).
>For details see: Lahey Fortran 90 Manual, App. B. As a consequence,
>using g77, one cannot process up-to-date programs got from scientific
>information exchange sources. And, of course, he/she will not be able
>writing up-to-date programs.
While this is fair comment about g77, I think the point that was being
made was that the availability of a free full Fortran 95 (i.e. g95) will
help a lot. The g95 project is still at a relatively early stage and
many people still don't appear to be aware of its existence. From what I
know of the effort required to develop commercial f90/f95 compilers, I
think the g95 developers have a bigger job ahead of them than they
realise, but I wish them well.
>- This compiler is distributed according to the GNU license. It means,
>everybody has the right to modify the source. Therefore, there is no
>guaranty on the quality of the compiler. As a consequence, this
>compiler must not be used solving problems requiring quality assurance.
>(On the other side, this condition pronounces the responsibility of
>the compiler vendors - the users have to pay for it.)
Agreed, though in practice it's actually fairly unusual to encounter a
'doctored' g77 of this sort. Quality-wise g77 is sufficiently good that
a large number of people do seem to be using it, with considerable
success. I must admit I've never found it to be up to the standard of
commercial offerings like Lahey F77L-EM/32 or Salford's FTN77, but we've
successfully used it to build substantial code on a variety of platforms
with little problem.
Ironically, I would say that g77 is actually part of the problem when it
comes to encouraging usage of modern versions of Fortran. Its free
status, wide availability and adequate quality mean that its in quite
wide usage. This is apparent from the amount of g77-related traffic on
the comp.lang.fortran newsgroup. Essentially, there is a sizeable user
base (particularly students) who will use just about any compiler if
it's free and easy to obtain, regardless of the long term cost of using
a limited and obsolete language standard.
Lawson Wakefield
! Interactive Software Services Ltd.
! The Fortran GUI specialists
! Winteracter home page : www.winteracter.com
----------------------------------------------------------
To unsubscribe, send to [address removed] the following
as the first and only line of the message body:
unsubscribe fortran
----------------------------------------------------------