Record length 32bit vs. 64bit in gfortran

[Warren Turkal]

On Friday 13 October 2006 15:38, Jerry DeLisle wrote:
> I think once in a while, somebody has to be bold and point the way and put
> up with the whining.  In another year, 64 bit computing is going to be the
> defacto for your hand held computer.  Sometimes people resist change for
> the better but we need to hold our ground.  We have a way to deal with the
> compatibility and it encourages people to let go of the old and get on with
> the new.

This seems right to me. The other option would be to get the unbuffered output 
standardized in the next version of the Fortran standards.

Speaking of incompatibilities, the incompatibilities of the .mod files is 
really annoying as well. I have systems that I have to support gfortran and 
pgf90, and I have to compile two sets of all the libraries to support them 
both. Where is the intervendor ABI like C++ has for Fortran?

wt
-- 
Warren Turkal, Research Associate III/Systems Administrator
Colorado State University, Dept. of Atmospheric Science