Post Reply 
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Speed-up for same instrument?
07-07-2008, 16:30
Post: #1
Speed-up for same instrument?
Is there a way to preserve the calculation at/near the place where it reports
"Convolving by detector pixel" but before "Painting input list"? My current
efforts to simulate Kepler require lots of images with slightly different
spacecraft pointing and noise, but with the same instrumental PSF. SkyMaker
seems to be spending most of its time computing the same PSF for each
simulation, and my guess is that things would go a lot faster if these
numbers could be recovered from a file before they are used.

Or maybe I am not following the calculation, or have missed the discussion
of the keyword that does this. If so, very sorry.

I am getting very useful results, so many thanks (again) for this
wonderful software.
Find all posts by this user
Quote this message in a reply
07-08-2008, 16:01
Post: #2
RE: Speed-up for same instrument?
Dear Dave,

Thanks for your comments! I've been using skymaker recently myself and it's a very useful piece of software.

I don't know too much about the internal details of skymaker, but I do know that it is multi-threaded; running the program on a machine with many cores should speed up a lot the 'painting object list' part (I don't know if the psf creation is multi-threaded).

To speed up the PSF computation, did you try changing the psf map size / oversampling?

I have seen your previous forum post concerning the psf files; my guess is that the external psf files are those produced by an as-yet unreleased piece of software, but they are standard fits files.

Cheers
henry
Visit this user's website Find all posts by this user
Quote this message in a reply
Post Reply 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)