Bugs: fake_pha

Bugs

1. Some fake_pha simulations have incorrect fit parameters and an incorrect number of counts.

2. fake_pha data set ID must correspond to a PHA spectrum.

3. get_source must be used to assign a previously assigned model to a fake_pha data set.

Bugs

1. Some fake_pha simulations have incorrect fit parameters and an incorrect number of counts.

   Multiplication of two integrated models, each with its own normalization, is not handled correctly. It results in incorrect fit parameters and an incorrect number of counts in fake_pha simulations.

2. fake_pha data set ID must correspond to a PHA spectrum.

   Sherpa 4.1 users must ensure that the data set ID supplied to fake_pha (S-Lang or Python help) must be associated with a PHA spectrum, either a real PHA data set or an empty PHA data set created with the dataspace1d (S-Lang or Python help) command.

3. get_source must be used to assign a previously assigned model to a fake_pha data set.

   To assign a model to a fake_pha data set which was has been previously assigned to another data set in the current Sherpa session - for example,

   python.sherpa> set_model(1, xspowerlaw.p1*xswabs.abs1)
   python.sherpa> set_model(3, get_source(1))
   

   where data set id=3 is the one to be used with fake_pha, then get_source (S-Lang or Python help) must be used to assign model id=1 to the faked data set, not get_model (S-Lang or Python help).