Hi,
I am trying to write a header file to import data to OpenDX. Positions
are in a FORTRAN output binary file. The first three records are
integers I, J, K followed by positions (float)
x_1,...,x_I,y_1,...,y_J,z_1,...,z_K. First I would like to be able to
read the positions. I wrote a xyz.general header that takes into account
that Fortran puts a byte at the beginning and at the end of each record
and that the first three numbers are integers, say:
1 2 2 2 1 1 (offset=9 bytes)
012 int int int 012 012 x_1 x_2 ...
I skip 9 bytes and I expect the pointer to be at the beginning of the
locations:
file = xyz.bin
grid = 128 x 128 x 32
format = msb ieee
interleaving = record
majority = record
header = bytes 9
field = locations
structure = scalar
type = float
dependency = positions
end
Well, the script doesn't complain but returns garbage. What confuse me
even more is that when I use "od xyz.bin -j 9 -tfF" I expect to see at
least the x-component printed correctly, instead I get numbers like
2.172357163208688e-153 or 8.804036109998351e+199. What is wrong in my
header? Data are in a different Fortran binary file with
interleaving=field. What is the best way to import these data without
converting them in different format?
I will appreciate any help.
Thanks,
Ted
