BUG: FOR4165 Warning Underflow of Smallest Positive REAL*8

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

SYMPTOMS

Compiling a source file that contains an assignment of the smallest positive REAL*8 literal constant to a REAL*8 or DOUBLE PRECISION variable causes the following compiler warning message:

warning FOR4165: underflow in REAL literal constant 2.225073858507201D-308

RESOLUTION

Use the TINY() intrinsic function to convert a number to the smallest positive REAL*8 value, and then use the result in your calculations.

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 and Workaround

C Compile options needed: none

      REAL*8 dx
      DOUBLE PRECISION dy
      dx = 2.225073858507201D-308  ! Warning - 0.000000000000000 assumed
      write (*,*) dx
      dy = tiny(1.d0)   ! work-around - dy = 2.225073858507201d-308
      write (*,*) dy
      END

Program Output

0.000000000000000E+000 2.225073858507201E-308