×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

Lost star when dithering causes capture module to hang

  • Posts: 421
  • Thank you received: 102
I've come across a couple related issues. The first issue is that when I set a large dither value (10 in this case), frequently Ekos will complain that it lost the guide star after a dither. It's like the dithering routine forgot to tell the guiding algorithm the new position of the guide star. It's still within the yellow box, not sure why it can't find it...

But this wouldn't be much of a problem, because it says "Lost track of the guide star. Searching for guide stars...", and it immediately finds the same star, and resumes guiding.

However, it doesn't notify the capture module that dithering is complete when this happens. The capture module stays stuck, saying "Dithering..." forever, even though dithering has long since stopped, and the guide module is happily guiding.

It's very frustrating to come back a couple hours later after starting a capture session, to see it only captured a single image and was stuck there.

I'm currently working around this by setting a smaller dither value (4 in this case), which seems to keep the guider from losing track of the guide star after a dither.
5 years 9 months ago #26522

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
Hi kross,

I would not recommend such a large dither. As you observed it can cause the guide star to be lost. But also, it can hurt your results when you go to align and stack the images. More of the resulting images will be cropped and you will end up with a smaller result. It might also make it slightly more difficult for the software to align the images, but I doubt that is much of an issue. So yeah the biggest problem is that it causes problems for keeping track of the guide star. If you really want to do it, could you make the tracking box larger?

Thanks,

Rob
5 years 9 months ago #26530

Please Log in or Create an account to join the conversation.

  • Posts: 1957
  • Thank you received: 420
I noticed that the objects I was imaging slowly moved out of the FOV when using a 12 pixel dither after each image. So I will decrease the number of pixels significantly, perhaps to 4 or so. But the guide star was never lost despite me using a 32 pix box.
5 years 9 months ago #26531

Please Log in or Create an account to join the conversation.

  • Posts: 421
  • Thank you received: 102
I'm going by the recommendation of Tony Hallas here


I confirm that a large dither really does help more with a DSLR than a small dither. I have no problems stacking, and I don't care if I lose a few pixels on the edge of the frame. My frame is 5000 pixels wide, and my object is in the center. If I lose a hundred on the edge, I don't care. I'd usually be cropping those out anyway.

Regardless, it doesn't change the fact that kstars capture module will hang when the guider temporarily loses, and successfully reacquires a guide star.
5 years 9 months ago #26543

Please Log in or Create an account to join the conversation.

  • Posts: 984
  • Thank you received: 160
What happens when the number of pixels for dithering is too low, has been discussed here:

indilib.org/forum/wish-list/3389-fixed-p...ue-to-dithering.html

You'll end up with a drift pattern that looks like "rain". The "rainy" picture was taken using 2,8 pixels dithering. The same picture taken using 10 pixels dithering did show perfect random noise.
5 years 9 months ago #26544

Please Log in or Create an account to join the conversation.

  • Posts: 984
  • Thank you received: 160
With regard to guiding I have experienced significant differences between mounts (or their respective drivers). A Losmandy G-11 works perfectly here with dithering set to 10 pixels whereas an LX200basic-driven Alt-5+FS2 (with all other equipment the same) regularly fails to even calibrate. The FS-2 + Alt-5 combination is of high quality and doesn't suffer from backlash or other imperfections that could explain the difficulties. The G-11 however never lost a guide star (neither at 2,8 pixels nor 10 pixels) and that's why I believe the root cause of your problem could possibly be found outside of the "dither code".
Last edit: 5 years 9 months ago by Alfred.
5 years 9 months ago #26545

Please Log in or Create an account to join the conversation.

  • Posts: 421
  • Thank you received: 102
Yes, it is certainly possible that the code told the mount to move a certain distance in this direction, and then the code is then looking for the guide star in that new position. However, maybe the mount didn't actually move as far as was instructed (or too far), hence ekos reports it has lost track of the star.

However, it does reacquire the guide star after reporting that it lost it, and it happily resumes guiding. Unfortunately, the capture module is stuck, waiting for dithering to complete, since it was never informed that dithering has, in fact, completed. This only happens when the guiding code loses and reacquires a guide star in the process of dithering.
Last edit: 5 years 9 months ago by Kevin Ross.
5 years 9 months ago #26546

Please Log in or Create an account to join the conversation.

  • Posts: 984
  • Thank you received: 160
Ok, so it is my understanding that we're confronted with two distinct problems. 1. The guide module losing the guide star during dithering which should not happen in the first place and 2. the capture module not being informed that guiding has resumed so it should continue taking pictures.
5 years 9 months ago #26547

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
Ok, I see. My frame size is much smaller than yours since I'm using a ccd with a couple thousand less pixels on a side. For a DSLR I could see a larger dither distance since you do have many more pixels, especially if it helps in another way. Are you guiding with PHD2 or Ekos Guide Module?
5 years 9 months ago #26548

Please Log in or Create an account to join the conversation.

  • Posts: 421
  • Thank you received: 102
Using ekos internal guiding
5 years 9 months ago #26549

Please Log in or Create an account to join the conversation.

  • Posts: 421
  • Thank you received: 102
5 years 9 months ago #26550

Please Log in or Create an account to join the conversation.

  • Posts: 2876
  • Thank you received: 809
Yes, I wrote the algorithm for KStars to continue guiding if PHD2 loses a star and then reacquires it, but not for the Internal Guider. That's a question for Jasem.

I will send him a message.

Thanks,

Rob
5 years 9 months ago #26552

Please Log in or Create an account to join the conversation.

Time to create page: 0.354 seconds