Mesh parameter in the KPOINTS - file

[serenade25]

Hello.
My name is S.Y Kim.
I'm a graduate student that theoretically research about physical properties of solar cells using by VASP 4.6.

I read the ppt that [Sampling the Brillouin-zone]
(Andreas EICHLER in University Wien).

The partial content of that is this :
「* mesh parameter
- determine the number of intersections in each direction
- longer axes in real-space
- shorter axes in k-space
- less intersections necessary for equally spaced mesh
* Consequences:
.....
for the direction corresponding to the long direction.
- typical values (never trust them!):
metals : (9x9x9)/atom
semiconductors : (4x4x4)/atom」

I can't understand about the sentence "never trust them!"

Is it means that input mesh parameter to 9x9x9 in metals and 4x4x4 in semiconductors? or never input it to that values?

I hope someone replies true meaning of that.
Thank you.
Kim.

[metosa]

It means that you should check the convergence of kpoints by yourself. Prepare a simple bash script and calculate the total energies for different kpoints meshes such as 4x4x4 5x5x5 6x6x6 7x7x7 8x8x8 9x9x9. Make a plot of kpoints vs energy and see what happens.....