BUG: /4fps1 and /Ox Compiler Options with Complex Calculations

• Microsoft Fortran PowerStation for Windows 95 and Windows NT, version 4.0

SYMPTOMS

Using the /4fps1 and /Ox compiler options together with source code containing complex number calculations causes the results of those calculations to be erroneous.

RESOLUTION

If you are performing complex number calculations, do not use the /4fps1 and /Ox compiler options together.

STATUS

Microsoft has confirmed this to be a bug in the Microsoft products listed at the beginning of this article. We are researching this problem and will post new information here in the Microsoft Knowledge Base as it becomes available.

MORE INFORMATION

Sample Code to Illustrate Problem

C Compile options needed: none

      COMPLEX A, B

      C = 3.1415926535897932384626434
      D = 3.0/2.0

      B = CMPLX(COS(1.5), SIN(D)) ** 3
      WRITE(*,*) B

      A = B - CMPLX(COS(4.5), -SIN(4.5 + C))
      WRITE(*,*) A

      END

Correct Output

    (-2.107958E-01,-9.775301E-01)
      (9.473733E-09,0.000000E+00)

Output using /4fps1 /Ox

     (-3.537882E-05,6.484614E-05)
      (2.107604E-01,9.775950E-01)