fxihub / spsim Goto Github PK

Hi,

I just tried to compile spsim with NFFT and run into errors in src/diffraction.c

There are a few typos where it says 'nnfft__' that likely should read 'nfft__'.

However, I corrected those on my source but I run into new errors:
/Users/Max/spsim/src/diffraction.c:613:4: error: no member named 'v' in 'nfft_plan'
p.v[k_3] = HKL_list[k_3]/max_v[0];
~ ^
/Users/Max/spsim/src/diffraction.c:614:4: error: no member named 'v' in 'nfft_plan'
p.v[k_3+1] = HKL_list[k_3+1]/max_v[1];
~ ^
/Users/Max/spsim/src/diffraction.c:615:4: error: no member named 'v' in 'nfft_plan'
p.v[k_3+2] = HKL_list[k_3+2]/max_v[2];

Those, I don't really can make out.

PS: it does compile without nfft...

I noticed that spsim does not work any more from the webpage. The reason was that libspimage was changed such that writing to file causes an error:

libspimage: warning: Unable to open noiseless_photon_out-00000.cxi in /root/src/libspimage/src/image_io.c:2224
libspimage: warning: Unable to open scattering_factors-00000.cxi in /root/src/libspimage/src/image_io.c:2224
libspimage: warning: Unable to open real_space-00000.cxi in /root/src/libspimage/src/image_io.c:2224

After I checked out libspimage commit a54d90a721bae71b4f1c22b1eba2e333b78243b9 and built spsim against that version everything worked again.

Z coordinates are often incorrectly interpreted when reading from PDB files.

Diffraction calculation in 3D using CUDA don't seem to work at the moment.
The image is squased in one dimension.

At line 304 of molecule.c, the rank of legal_atom_names should be 2*(mol->atomic_number[i]-1), not mol->atomic_number[i]-1, since each atom is represented by 2 characters.

diffraction.c: calculate_pixel_solid_angle(CCD * det)

There should be a way that allows for parsing PDB files with and without having the element columns (77-78), some kind of hybrid between the code before and after commit fc920d8

When using the function get_Molecule_from_pdb for PDB entry 1GRO the function wrongly identifies some Hydrogens for Heliums. We are using the atom entry (columns 13-16). These can be modified and element assignment may be ambiguous. I suggest to use the information of the element entry (columns 77-78) instead.