I am finding more problems with CVF and XP. I am trying to go back to default real*8. This shows up in the compilation options window as "real_size:64". When examining reals in debug mode, they are all single precision except for ones that I have explictly declared as real*8. I've tried deleting all the modules, object files, and executable from the debug directory, but a new compile still ignores the "real_size:64" option. Any thoughts on how to get this to work again? I'm now wondering what other options are being ignored.
For more complete information about compiler optimizations, see our Optimization Notice.