Validation: a fundamental question

Not a guru

hello, i am new cfd beginner and i did some simulation of flow around cylinder at various Reynolds in fluent and cfx to get started. I did mesh dependency study and solution is not varying.

when post-processing the global Cd = f(Re) I obtain for both code the well know curve. I have found an example on the web below:

http://imartinez.etsin.upm.es/bk3/c12/Forced%20and%20natural%20convection_archivos/image022.jpg

Now if I also calculate the same local Cd for each nodes or face around the cylinder as a function of local Re (Re(theta)) I obtain the following:

- with cfx every point of each node or edge are located on this same curve (and the integrated average is). - with fluen the points are not on the curve but the integrated average is.

which code is right, fluent or cfx?

should friction versus Re follow this curve both globally AND locally or only the global value.

Later, I plan do look at the Nusselt as well (global and local)

What do you find with the software you use?

thanks in advance.