Re: Difficulty with Relatively Simple IF Structure

On Nov 22, 7:11 pm, "robin" <robi...@xxxxxxxxxxx> wrote:
"monir" <mon...@xxxxxxxxxxxx> wrote in message

news:a1c21aaa-c40d-4737-8141-89bc8a2e3f17@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
On Nov 18, 9:46 pm, e p chandler <e...@xxxxxxxx> wrote:

4) I've used print statements in the subroutine for debugging
purposes, and not the full debugging and checking options available in
g95 compiler

Why not, after e. p. chandler suggested that you do?

5) I've all the correct declaration in the FULL version of the
subroutine.

How do you know?
Do you have IMPLICIT NONE in this subroutine?
It's not shown here.


Thank you all for your interest, and I apologise for the delay in
responding.

1) As I indicated earlier, the program has been extensively tested and
works perfectly for other scenarios where that particular nested IF
block is not invoked in Subroutine SHTSBD2.

2) The numerical method works perfectly as well for all other tested
smooth, monotonic, integrable and well-defined 1D functions over the
XR-range-of-interest.

3) The difficulty appears to be "conceptual" rather than programming
or numerical. The current problem situation is somewhat analogous
(albeit a weak one!) to specifying X1 and X2 believed to bracket a
root of a function, and it turned out to be a false bracket. In such
situations, one can't blame the algorithm!!

4) I'm currently testing an alternative concept using Excel VBA. The
results so far are very promising.

5) Once the extensive numerical experimentation is done, will return
to the Fortran "modified" nested IF structure in Sub SHTSBD2 where
your help would be greatly appreciated. Will focus only on the IF
block, variables names, initial values, etc.

Regards.
Monir
.