Knowledge Base

FIX: Debugger displays Floating Point Notation for REALs

PSS ID Number: 105936

Article Last Modified on 3/24/2000

The information in this article applies to:

Microsoft FORTRAN PowerStation for MS-DOS 1.0

This article was previously published under Q105936

SYMPTOMS

The debugger will display real values only in floating point notation in the Locals and Watch windows. Other modes of display, such as scientific and exponential notations are not available.

STATUS

Microsoft has confirmed this to be a problem in FORTRAN PowerStation version 1.0. This problem has been resolved with FORTRAN PowerStation maintenance release version 1.0a for MS-DOS.

FORTRAN PowerStation version 1.0 can be differentiated from the maintenance release version 1.0a by invoking the linker. Typing link32 | more from \F32\BIN directory will show version 2.8 for FORTRAN PowerStation version 1.0, and it will show version 1.0f for the maintenance release version 1.0a.

MORE INFORMATION

Many numbers are poorly represented in floating point notation. For instance, a very small real value may be displayed as 0.

Sample Code

C Compile options needed: /Zi
      real*4 x,y
      x = 1.2345E38
      y = 1.1234E-38
      print*, x,y
      end

In the debugger, step through the program until the print statement. Local and watch window on x and y will display:

X = 123450001585154100000000000000000000000.000000
Y = 0.000000

Additional query words: 1.00 IDE

Keywords: kbbug KB105936
Technology: kbAudDeveloper kbFORTRANPower100DOS kbFortranSearch kbZNotKeyword3