×

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

Bi-monthly release with minor bug fixes and improvements

New horizon map drawing errors

  • Posts: 1009
  • Thank you received: 133
With joy I noticed that the Artificial Horizon now properly updates also in the Ra-Dec mode 
However, it now has a drawing error (both in Ra-Dec and Alt-Az modes) if features of the map "go backwards in azimuth" - see the attached image.  It's a tower with a larger platform on top, so there are points where Az decreases compared to the preceeding ones.  Here's an part of the Az,Alt values in that area:
     237.900      14.0500
      249.600      12.8500
      256.700      12.1000
      256.300      32.0500
      248.550      33.7000
      248.300      37.1000
      263.650      48.2000
      279.450      38.8500
      279.750      0.00000
      270.000      0.00000
It was rendered properly in the old version.
 
2 years 11 months ago #70630
Attachments:

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

  • Posts: 1208
  • Thank you received: 559
Peter,

Sorry, I don't fully understand what the issue is, but would be happy to look at it. Could you please spell it out for me--e.g. send me the full data, and say (a) what happens, and (b) what you would have expected to happen?

It wouldn't shock me if my recent changes caused older artificial horizons to regress. The good news is I made it easier to edit. Can you also try first? You can go to settings -> artificial horizon and you should see your artificial horizon on the sky map, with red points on your stored coordinates and white lines between them. You cannot move those points on the sky map, nor click on them--you can just see where they are (editing them is a future, hopefully). However, if you have the points displayed, and then look at the list of points, and select one of the text points with the mouse, (and can move between them using the up/down arrow keys), you should be able to see the current text point highlighted on the sky map (it's dot is larger).  You should then be able to delete or edit the coordinates, or insert new points, using the text input and buttons near it.

BTW, previously the artificial horizon points defined a polygon. This is no longer the case. Right now they define a horizon. That is, if you have 2 values at the same azimuth, only the higher elevation one matters. The altitude values between two points are the interpolated values of those points. So, if the "underneath" points are causing a problem, you can probably delete them. Warning, though, there is no "undo" at this point.

Hy
2 years 11 months ago #70668

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

  • Posts: 1009
  • Thank you received: 133
Hello Hy,
I've attached a screenshot taken with an older version of kstars that shows how the horizon map actually should look. In the area of the overhang it is painting things twice it seems. But your further text explains why:
This of course is completely different from how it worked before. You cannot mark general obstructions anymore, as they often do not follow this linear approach and have overhangs. So I think this constraint brings a substantial reduction in usefulness of the horizon map 

Edit: Forgot the image....
 
Last edit: 2 years 11 months ago by Peter Sütterlin.
2 years 11 months ago #70688
Attachments:

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

  • Posts: 1208
  • Thank you received: 559
Sorry about that Pit. My hope is to add scheduler integration, so that the scheduler is aware of obstructions, to compensate for that loss. Also the previous implementation had issues with knowing which side was the interior (vs exterior) of the polygon--there were cases where the polygon was interpreted wrong and the sky was colored red! To fix all this, and to facilitate scheduler integration, it made sense to make the artificial horizon a "horizon" (as opposed to a polygon). It is possible that a future addition might include a maximum line, e.g. to indicate that nothing above a line is visible and shouldn't be scheduled. Don't know, though how possible it would be to have multiple minima and maxima, like your image shows (for scheduler integration). 
The following user(s) said Thank You: Peter Sütterlin
2 years 11 months ago #70721

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

  • Posts: 460
  • Thank you received: 69
One other option, that is easy to suggest if I'm not actually contemplating doing the work myself, is to use the minimum vertical position of the alpha channel in the image at say 5 deg increments horizontally to build an initial polygon set of points that could be perhaps tweaked later to add more points "by hand" to accommodate outcrops.
2 years 11 months ago #70725

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

  • Posts: 1009
  • Thank you received: 133
No worries, Hy.  Sometimes a step back is needed to proceed 
I am sort of afraid that it would definitely be needed to be useful in the Scheduler.  At least, I find it hard to believe that for most people the obstruction curve would be monotonic in Az without either losing actually unobstructed area or leaving obstructions unhandled....
Last edit: 2 years 11 months ago by Peter Sütterlin.
2 years 11 months ago #70734

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

Time to create page: 0.617 seconds