*** Please do not post license keys - please report any problems with licensing to 'admin' by private message ***
Please include the following details in any bug report:
* Version of SharpCap
* Camera and other hardware being user
* Operating system version
* Contents of the SharpCap log after the problem has occurred.
[If SharpCap crashes, please send the bug report when prompted instead of including the log]
Hi Robin,
Not sure if this is a bug or just a weird artifact stacking effect, I get this develop through the course of the live stack always to the right side of the image so initially thought sensor tilt but having looked at the individual sub fits they are fine.
This is a grab of the drift tab this is after 189mins of 3 min subs so 63 stacked subs dithered every 4 frames unless I refocused the frame to frame pixel offset is always in a similar range to that in the screenshot which I think is reasonable unless I am misunderstanding the value this scope is seeing at 2.2 arc seconds a pixel.
Drift Tab Live Stack
Drift.png (23.89 KiB) Viewed 1393 times
This is the lower right corner at 100 percent zoom from the live stack
live stack after 63 subs 100 percent zoom
Sharpcap_100.png (348.17 KiB) Viewed 1393 times
This is the same corner from one of the sub fits files they all look like this other than the dither movement. slightly square hence dither for drizzling later.
definitely an odd one. I can't quite see what would cause that exactly, but will go through some ideas (and why they don't quite fit) just in case somewhere along the way I hit on something useful...
To me, that looks like there has been one or more frames stacked with an incorrect offset, so that you have got second copies of all the stars offset by a few pixels from the main copy. However, I'm not sure how that could only apply in the bottom right of the image (by the way - is the onset in the bottom right corner abrupt or gradual as you move further into that corner).
On the same theme, if you had something move during one of the frames, you would possibly get 2 copies of each star in that frame, and if that frame was added to the stack then it might lead to something like this across the whole frame. But, that again would apply to the whole frame and you would see the problem in one of the subs...
Maybe something is shifting internal to the optics of the telescope as the telescope tracks and the weight distribution changes - potentially that could cause the image to change in odd ways, perhaps with a stretching in some direction. You could measure the pixel distance between a bright star top left and one bottom right in the first and last frames to see if they are the same (we would expect that distance to be constant even with drift/dithering) - if that changes significantly something odd is going on.
Do you see the same effect if you restack the images using the SharpCap folder monitor camera? How about in a different stacking application?
Hi Robin,
I will go through the other things in your list, but the image below is the same corner in pixinsight 135 frames stacked no other adjustment other than stretched to see stars, so think it is probably not an issue with the optics flex. Also, the images posted were the bottom right but the whole rightside actually shows the effect and it is more pronounced the further right you go so as you go back left it improves and by the time you are at say 20% across the image back towards the left it is imperceptible.
135 frames stacked_proc.jpg (421.49 KiB) Viewed 1349 times
hmm, OK, in that case maybe it is possible that somewhere during the live stacking process the alignment calculation comes up with a result that has a scale factor in it, rather than no scaling. That could mean that one or more frames get stretched before stacking, leading to a drift of alignment across the frame.
If you can repro the issue in the folder monitor camera then I could probably debug it if you could share the files.
Hi Robin,
Have used the folder camera and gone through each frame as a single image looking at the bottom right and all the stars are round and the only movement I see is on a dither. I then did the same progressively stacking the subs into the live view and the effect starts to manifest in there and the more frames stacked the more it shows up I am quite happy to give you the data if you want to have a look, but they are fits files so quite large you only need 25 to see the effect and that's a 375mb rar file so I could lob it up on WeTransfer and message you a link to the file if that is okay then you could see what I see hopefully.
happy to have a look at the data if you upload it, but before that, it might be worth checking the Live Stacking alignment settings in SharpCap. In particular check that you are getting a decent number of stars detected (20+) in each frame and that you have not somehow got the 'align on a comet' setting switched on.
Hi Robin,
Only star alignment is on, I also activated the show alignment stars option and watched it through the first 25 frames, and it is using the same stars for all frames so will upload the files and send you a link to download them to see if you spot something. If it was an issue with the actual scope stretching the image I would have expected to see the same issue in PixInsight when I integrate the subs but they integrate fine so currently somewhat bemused as to the reason.
thanks fo sharing - I have downloaded and can confirm that I am seeing the same problem in the bottom right corner of the stack, but nothing wrong in the individual frames as far as I can see, or in the top left. I will dig deeper into what is going on and report back.
OK, I understand what is wrong now, but it's weird... The code hasn't changed for ages, and if it is wrong now then we should have had problems with live stack stars all over the place - derotation of Alt/Az images should have failed too, but they have worked.
I suspect that the code that looks wrong now was actually once right - something, somewhere else in the program behaved differently but was compatible with the current code so everything worked. Then that other thing got changed and now we have a mismatch and things break. I am still trying to track down exactly what the other thing that changed was (and when) to understand if there are any other potential issues that I should check.
