For a single 1000 m depth profile we go from a ~30 MB p file, to a ~110 MB p2mat file to a ~ 500 MB profile file... a whopping order of magnitude increase in storage requirements. 0.5 GB per profile is too much. I think that saving the p2mat file makes a lot of sense because it saves time loading the data, but we should not save all windowed shear / acceleration spectra.
For a single 1000 m depth profile we go from a ~30 MB p file, to a ~110 MB p2mat file to a ~ 500 MB profile file... a whopping order of magnitude increase in storage requirements. 0.5 GB per profile is too much. I think that saving the p2mat file makes a lot of sense because it saves time loading the data, but we should not save all windowed shear / acceleration spectra.