AIPS Alpha (OSF/1) Install Notes NRAO

This page updated on $Date: 1997/05/21 18:05:43 $

The floating point model on these systems can cause problems; usually these are when bad things such as floating point overflows or underflows occur which may indicate bad data or a programming error/oversight. If you run into one of these, and are fortunate enough to have a second AIPS system on non-Alpha hardware, please see if you can reproduce the error on it. If you cannot, try and eliminate the possibility that it's your data that is causing the problem before proceeding. Finally, consider rebuilding the task in debug mode so that you can get a traceback and identify exactly where the error is occuring (doing this is not easy).

The DEC Fortran Compiler emits some rather grim informational messages when compiling hte PP.FOR and UNSHR.FOR utility programs. However, these can be ignored.

Patrick P. Murphy, Ph.D.
work / home / comments