Compaq Visual FORTRAN on Vista (64 bit)?

Compaq Visual FORTRAN on Vista (64 bit)?

I have tried to install Compaq Visual FORTRAN on Vista (64 bit). But Developer Studio does not start. Only an error message is writen that program is terminated. Is there any chance to make it working?

7 posts / 0 new
Last post
For more complete information about compiler optimizations, see our Optimization Notice.

Try right clicking on the shortcut for Developer Studio and select "Run as Administrator".

Wouldn't you rather use a compiler that generates 64-bit code?

Retired 12/31/2016

Steve thank you very much. It works. I have IVF 9.1.034 + VS2005. But the problem is thatsometimesI have to return to older CVF projects . GenerallyCVF projects can be converted to IVFeasily. But there are also cases where are problems. Even when conversion is successful results of calculation can be different.I have also noticed very strange thing that each compiler interprets the code in different way. For example original code (part of very huge FORTRAN code) has following "bug":

DATA RFLUX/1.07E+06,1.74E+08,3.64E+07,8.53E+07/
!.. Fluxes for SEE V7 from Woods in August 2004
DATA RFLUX/2.4542E+06,1.9941E+08,4.1809E+07,9.7841E+07/
!.. SEE fluxes at F10.7=80 (i.e. divided by 3)
!.. DATA RFLUX/3.60E+05,5.80E+07,1.21E+07,2.84E+07/
! F74113 Auger fluxes at F10.7=80
DATA RFLUX/3.60E+05,1.00E+06,8.00E+06,3.55E+06/

Three the same statements DATA RFLUX!

No error messages during compilation and CVF took the last one. But after porting to IVF, IVFtook thefirst one! No wonder that results of calculation were different. But to find thiserror it took several days!


I am trying to use my old compaq visual fortran on my Intel fortran 11.1, but it cannot run. I got a message saying that certain files, such as mt.exe, cannot be found. Can you give me a hint? Thanks

The draft of the Fortran standard states in Section 5.2.5: A variable, or part of a variable, shall not be explicitly initialized more than once in a program.

Other compilers may have switches to pinpoint such errors. For example, gfortran -pedantic would tell you:Warning: Extension: re-initialization of 'x' at (1).

Choosing compilers on the basis of how they treat non-conforming code certainly has its pitfalls. One of these is the distinct possibility that a revised compiler may behave differently than an older version!

What do you mean by something along the lines of "I am using on my "?

If you meant to state that source code that you could formerly compile with CVF cannot be successfully compiled with IVF 11.1, please state that explicitly, give a specific example and attach the error messages.

If the compilation goes through but the resulting .exe file does not run, that is a different matter.

By distinguishing between error messages from (i) compiler (ii) linker and (iii) runtime, you improve your chances of obtaining a resolution.

The manifest tool mt.exe is included in the Windows SDK. Current versions of the Microsoft linker have a /MANIFEST:NO option, which may be worth looking at if you do not have the Windows SDK.

Thanks. I meant the the project which runs well in compaq cannot be built in Intel. I guess the problem occured because I did not extract the compaq project correctly to intel.

Leave a Comment

Please sign in to add a comment. Not a member? Join today