Re: Random number and fortran standard



--M-->> <no.spma@xxxxxx> wrote:

The executable made by gfortran give every time the same number in output.

With the executable made by g95 I obtain a different number for each run.

So, who is the correct compiler?

Both. The standard allows either behavior.

--
Richard Maine | Good judgment comes from experience;
email: last name at domain . net | experience comes from bad judgment.
domain: summertriangle | -- Mark Twain
.



Relevant Pages

  • Re: Fortran Version Question Apr. 8, 2009
    ... gfortran supports ISO C Binding as does g95. ... gfortran has a rich set of intrinsics procedures ... Note the Standard ...
    (comp.lang.fortran)
  • Re: Hello World...
    ... > research into the code I typed I found that there was a standard so I ... > me the warning, but I also have a copy of Borland Turbo 1.01 and this old ... a completely correct compiler or a completely correct book ... or one of earlier editions of "The C++ Programming Language" to see what ...
    (comp.lang.cpp)
  • Re: The "F" language.
    ... I was briefly a member of the committee developing a different standard, ... get something through the standard's process and into compilers. ... You could start with, say, gfortran, and do it there. ... Or g95 if Andy or someone is still around to work on it. ...
    (comp.lang.fortran)
  • Re: repeated namelist behavior
    ... Steve Lionel wrote: ... and merge all the namelists? ... if the standard actually nailed this down so we ... could count on the behavior we observed with g95. ...
    (comp.lang.fortran)
  • Re: Default initialization and save
    ... My question is if g95 is being more picky about the standard, ... "more picky" about the standard, but that would be an understatement. ... The correct statement would be that any compiler that cannot diagnose ...
    (comp.lang.fortran)