[QE-users] Non consistent C->P, P->C cell transformation generated by Quantum Espresso

[Michal Husak]

Hi

When the input to QE is done through space group number, the centered 
space groups (C,I,F) in conventional
cells are automatically transferred by QE to primitive cell P (to 
speed up the calculation) ...

We need to transfer the QE result back to the original conventional 
cell (to be able to compare the data from standard CIF
file with QE results) ...

I will be easy if QE will use always the same matrix for C->P 
transformation ...
Unfortunately this is not the case QE use different matrix like:
0.5 -0.5 0
0.5 0.5 0
0 0 1
than it use
0.5 0.5 0
0.5 -0.5 0
0 0 1
e.t.c. ....

Is there any way how to easy decipher form QE output what 
transformation was done with the input data (what matrix did QE chosen
for the given C->P transformation) ?
It is a pity QE does not list the transformation was done, nor what 
matrix was chosen ...

Can this be done e.g. by calculating the transformation matrix 
between the P cell parameters and original C conventional parameters
(we hope at lest the transfer of cell to Cartesian vector is done 
always identically) ?

Michal Husak