Analyzing takes a long time
Stacking program, astrometric (plate) solver, fits file viewer
Brought to you by:
han59
Menu
▾
▴
I’ve loaded about 300 images to stack into ASTAP, each image is taking over a minute to analyze. Is there a setting that has been changed, or something I can do to speed things up?
That is weird. Can you try it on some old images? If the old image are analysed quickly, can you provide me with two of those images for testing?
Han
I am still in the middle of doing the analyze that’s taking a long time. I’ve noticed that images are now taking about 30 seconds to analyze. These images were taking over several nights. The really slow ones seem to be on one specific evening. Is 30 seconds closer to what I should expect?
Normally maybe 5 seconds maximum. If there are no stars (clouds) on the image, it could slow the process. Are the image good? Also the conversion (once) )from raw to FITS could take a few seconds. What are the image dimension or how many megapixels?
They are 40mb fits files from a OSC QHY183c camera 5544x3684. I think one
of the evenings my guiding wasn’t very good. And I know another evening
there were some clouds coming in and out. When I look at the separate image
files, some of them have a cross pattern on them. Kind of looks like a
checkerboard or a tartan. But they are all slow. Nothing takes 5 seconds.
I’ll try to upload a couple of examples to Google drive and send you a link.
On Wed, Nov 25, 2020 at 2:40 PM han.k han59@users.sourceforge.net wrote:
Last edit: han.k 2020-11-25
It looks like a checkerboard because it is a raw image. 5544x3684 should normally not take 30 seconds.
Han,
Here is the link to 2 of the fits files.
https://drive.google.com/drive/folders/1EL2vRiyrzLnctmcpTpW-XZJTs8NOcuy6?usp=sharing
On Wed, Nov 25, 2020 at 5:05 PM han.k han59@users.sourceforge.net wrote:
That is horrible slow. If I flip them horizontal it goes quick. Must be something weird happening with the background calculation. Will try to find a fix.
Found the fix. Download and install this development version 0.9.457:
http://www.hnsky.org/astap_setup.exe
The problem was in the background noise calculation. For this it takes a selection of pixels at a specific step size. The step size was even value resulting that only red or green pixels where measured. I added an extra check for an odd step size so that all colour pixels are sampled.
Clear skies, Han
Awesome, thanks!
On Thu, Nov 26, 2020 at 3:10 PM han.k han59@users.sourceforge.net wrote: