When we generate grids using https://github.com/NNPDF/runcards with a proper dynamic scale choice - that means a scale prescription that has more than one value for each subgrid - we keep the full scale information in the optimized subgrids. This is not a problem as far as storage is concerned, but we can not see which scale values are 'empty' and which are 'filled'. This is a problem
if we want to reconstruct the dynamic scale choice (see for instance https://github.com/NNPDF/runcards/pull/61#issuecomment-1048757526) through its range: a grid generated with the default parameters will always show all 40 grid points in muf as ranging from 10 GeV to 10 TeV.
It will also potentially and unnecessarily slow down the evolution of those grids.
When we generate grids using https://github.com/NNPDF/runcards with a proper dynamic scale choice - that means a scale prescription that has more than one value for each subgrid - we keep the full scale information in the optimized subgrids. This is not a problem as far as storage is concerned, but we can not see which scale values are 'empty' and which are 'filled'. This is a problem