OMM.LongOutput not honored

[Alberto Garcia]

(Reported by Marcos Verissimo)

1) There seems to be a small bug in one of the variables, namely OMM.LongOutput. I have set it to false and still siesta outputs all the info on the band iterations:

(...)
+---------------------------------------------+
| k point 2067 of 2067 |
+---------------------------------------------+
| minim: icg = 11 |
| minim: 2*Tr[(2*I-S)*S] = 16.0000000 |
+---------------------------------------------+

[Fabiano Corsetti (f-corsetti)]

Hi Marcos,

That is the correct behaviour of OMM.LongOutput. You should find that it prints out a lot more information if you set it to true!

If you would like to suppress all output from OMM you can apply the attached patch.

Fabiano

[Alberto Garcia (albertog)]

Thanks, Fabiano.

Your patch will suppress all output when using the flag. Is this something that can be committed without impairing
functionality (say, the ability to see what is going on at some intermediate level, without being snowed over?).

[Fabiano Corsetti (f-corsetti)]

Hi Alberto,

I'm not sure. The way I originally coded it, OMM.LongOutput = .false. prints a few lines of summary info for each k-point at every SCF iteration (while .true. prints a line for every iteration of the OMM algorithm). As Marcos found, the problem with this setup is that even .false. ends up printing a lot of information when you have many k-points.

Maybe it makes sense to apply the patch, so that users can choose between no information and full information. Then .false. basically gives you the same output as a standard diagonalization run.