Re: isis-1.0.8 is available

Craig Heinke <cheinke_at_email.domain.hidden>That's something to think about.  I'm finding that each additional parameter I 
>let float increases the fitting time substantially, so I'm looking for 
>parameters I can fix without a great impact on the fits.  In my case I'm
>dealing with pileup of under 10%, and extracted counts in a 4 pixel radius 
>around an on-axis point source.  Most of my pileup will be in the 2-photon 
>case, due to the low count rates, leading to (to first order) one 

Does "print_kernel (1);" produce results that are consistent with your
expectations?

>energy-shifted spectrum of piled photons.  Two variable parameters determine 
>its normalization--the alpha parameter and the psffrac parameter.  As the 
>science doesn't actually depend on either of those values directly, it seems 
>to me (correct me if I'm wrong) that allowing only alpha to vary will still 
>correctly normalize the first-order pileup.  So even if psffrac is between 0.9

What values for alpha are you seeing?

>and 0.95, I shouldn't go too far wrong in my spectral fits in freezing it if I
>let alpha vary.

How do you parameters and alpha change when you use psffrac=0.92
instead of 0.95?

Thanks,
--John
----
You received this message because you are
subscribed to the isis-users list.
To unsubscribe, send a message to
isis-users-request_at_email.domain.hiddenwith the first line of the message as:
unsubscribe