Aneta Siemiginowska

> In some sense, they allow the analysis of PHA data without an ARR and RMF. For observed data, that should not happen since we always have ARFs and RMFs;...

@DougBurke it looks like this is crates issue. I can read using `astropy` ``` In [1]: from sherpa.astro.ui import * In [2]: ls 70024000_sis0.arf 70024000_sis0_bgd.pha asca.tar.gz 70024000_sis0.rmf 70024000_sis0_src_20.pha test_none.pha In...

``` In [5]: load_pha('70024000_sis0_src_20.pha') --------------------------------------------------------------------------- RuntimeError Traceback (most recent call last) in ----> 1 load_pha('70024000_sis0_src_20.pha') in load_pha(id, arg, use_errors) /anaconda3/envs/ciao-4.13/lib/python3.8/site-packages/sherpa/astro/ui/utils.py in load_pha(self, id, arg, use_errors) 1899 id, arg = arg,...

@DougBurke I just sent the error msg. Do you think this is the issue in the sherpa backend or directly in crates?

CIAOX still shows this issue.

@DougBurke is this still an issue? I do not get an error in the 4.12.1 candidate release in ciao.

@DougBurke yes, I see it with this example.

@hamogu Is there any difference between set_arf/rmf and the proposed change to load_arf/rmf?

Just a comment which may be useful (or not) for the discussion here. The original idea with `set_arf/set_rmf` was to be able to have several RMF/ARFs available and then select...

At some point there must have been a check that the independent axis are sorted - ascending/descending. The models are defined on the ascending axis. The data and the model...