PushTo direction

[Gfischer]

Robin,
Just wanted to make sure I was getting at least 100 stars. I have had some instances when I get a solve fail . I let it fail a couple times then move scope a degree or so and then get a solve. I tried when a position wouldn’t solve for a few times, switched to ansvr and it then solved. Not very often though.
George

[admin]

Hi George,

if you get places that fail, I'd be grateful if you could snapshot an image in FITS format and share it with me. I can then use those images to try to understand why failures are happening and see if I can improve.

cheers,

Robin

[Gfischer]

Will do. I noticed a folder with plate solve images being saved in sc . Are these failed images? Thought you might have set it up for trouble shooting.

George

[admin]

Hi George,

the folder of plate solve images is created when you have turned on the 'additional logging' option for plate solving. While that option is on, it will save *every* image sent to the plate solver along with all the related information. I suspect you had that option on once and have since turned it off again , but the folder still exists.

cheers,

Robin

[Gfischer]

Robin,
I had problems with sharpsolve last night it wouldn't solve and failed several times. I attached the log file from the fail and another log file after switching back to ansvr which solved with the same exposure and gain. I tried to upload tiff but it was to big. Link on bottom.

Hope this helps
George

https://www.dropbox.com/scl/fi/tv3x06i6 ... 6ra8i&dl=0

[admin]

Hi George,

thanks for sharing that file and the logs - I am going to try to look into the cause later this week.

Robin

[admin]

Hi George,

I think I understand the issue now - there is a lot of skyglow noise in that image (particularly in the red channel). That impacts on the detection of stars in the image, and although 100 or more are detected, the quality of the detection is poor.

If I re-load the image into the SharpCap test camera (set RAW16 colour space, Force BGGR in Debayer Preview to get correct colours), it fails to solve. If I then turn on background subtraction in the SharpCap pre-processing, it works, since the significant reduction in the background brightness helps improve the star detection.

I will try to work out a way to include something like that in the automated star detection that I am using for SharpSolve.

cheers,

Robin

[Gfischer]

Robin,
Wow your reply was fast. I had a suspicion there may have been trouble with background. Your solver worked great before but this time there was a near full moon out. I notice during full moon nights my frames when live stacking tend to be red balanced. I’ll apply background subtraction to my profile for plate solving maybe tonight.
I’ll let you know if that works.

George

[Gfischer]

https://www.dropbox.com/scl/fi/26she2b8 ... o106p&dl=0
https://www.dropbox.com/scl/fi/26she2b8 ... o106p&dl=0

Robin,
I added background subtraction and it helped a lot. I attached 2 files, the Raw16 file failed solve. I then changed to Mono8 and it solved ,no exposure or gain changes.
Something else I noted. When a solve completes the distance to target is many times, larger than the radius of search display. What about using distance to target for search radius until you are getting close (maybe 10 deg) then fixed. I noticed that when radius is smaller than distance to target you get sequential fails until radius goes back to 180.

George

[admin]

Hi George,

both links in the post above download exactly the same file - one that does solve for me. I suspect that somehow you ended up with a duplicate of the link for one file.

A nice idea about using the distance to move as a guide for the plate solve radius, although I wonder if at large distances (more than 60 degrees or so), it might cause too much of a slowdown? I will experiment a little.

cheers,

Robin