fcom/fcompp and fstatsw different results

The FCOM, FCOMPP, and FSTATSW are three Fortran compiler directives used to control the floating-point precision and error handling in a Fortran program. FCOM and FCOMPP are used to set the precision of the floating-point calculations, while FSTATSW is used to control how floating-point errors are handled.

If you are seeing different results when using these directives, it is likely due to the differences in how they are configured.

Here is a brief overview of each directive:

• FCOM: This directive sets the precision of the floating-point calculations to either single or double precision. Single precision uses 32 bits to represent floating-point numbers, while double precision uses 64 bits. If your program requires high precision calculations, you may want to use double precision.

• FCOMPP: This directive sets the precision of the floating-point calculations to extended precision. Extended precision uses 80 bits to represent floating-point numbers and is often used for scientific or engineering calculations that require high accuracy.

• FSTATSW: This directive controls how floating-point errors are handled. By default, Fortran programs will continue executing even if a floating-point error occurs. However, if you set FSTATSW to "TRACEBACK", the program will stop executing and display a traceback of the error message.

If you are seeing different results when using these directives, you may want to check the compiler options or command-line arguments that are being used when compiling your program. Different compiler options may result in different precision or error handling settings, which can affect the results of your program.

It's also possible that the differences in results are due to subtle differences in the input data or algorithm used in your program, which can result in different floating-point calculations or errors. In this case, it may be necessary to carefully review your code and input data to determine the cause of the discrepancies.