MPB Users:
I would like to thank Mischa Megens and Steven for their helpful
replies to my previous question. Once I removed hdf5-1.8.1 and
recompiled the older hdf5-1.6.7 libraries, both h5utils and the recent
version of mpb installed without a hiccup.
My question today is about the epsilon.h5 files that MPB uses to store
the dielectric data. I have used an HDFviewer application and
Matlab's hdf5read function to rip apart the file and pull out the
dielectric 'slices'. This is the first time that I have used the hdf5
file format, but from what I can gather, data is stored in a
hierarchical structure (similar to a UNIX directory tree) that allows
each set of data to be separated and organized in a logical fashion.
However, I am having some difficulty making sense of how the data is
stored. From what I can see, the data is stored like this:
epsilon.h5 --> data
description
epsilon.xx
epsilon.xy
epsilon.xz
epsilon.yy
.
.
.
epsilon_inverse.xx
epsilon_inverse.xy
.
.
.
lattice vectors
The fields "description" and "lattice vectors" are straightforward
enough. My problem comes from making sense of the "data" and
"epsilon" data sets. Each of these sets forms a 1x64 row. When I
force the data into a 8x8 matrix, I can generate an image that roughly
agrees with my dielectric function.
So my questions are:
1) When I open the data from epsilon.nm , what data slice am I
actually looking at?
2) How does h5topng grab all of the data points for a given slice (say
all points where x = 15)
True Merrill
251.648.4093
[EMAIL PROTECTED]
_______________________________________________
mpb-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/mpb-discuss
