×

INDI Library v1.9.6 Released (21 May 2022)

Bi-monthly INDI Library released with new drivers and bug fixes.

Internal Solver stopped working

  • Posts: 41
  • Thank you received: 5
<code>2022-03-21T23:08:35 Solver Failed.
2022-03-21T23:08:35 Solver was aborted, timed out, or failed, so no solution was found
2022-03-21T23:08:34 Starting Internal StellarSolver Astrometry.net based Engine with the 4-SmallScaleSolving profile. . .
2022-03-21T23:08:34 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
2022-03-21T23:08:34 Image width 2028 pixels; arcsec per pixel range: 0.588521 to 1.17781
2022-03-21T23:08:34 Scale range: 19.892 to 39.81 arcmin wide
2022-03-21T23:08:34 Configuring StellarSolver
2022-03-21T23:08:34 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
2022-03-21T23:08:34 Stars Found after Filtering: 15
2022-03-21T23:08:34 Keeping just the 15 brightest stars
2022-03-21T23:08:34 Removing the stars with a/b ratios greater than 1.5
2022-03-21T23:08:34 Stars Found before Filtering: 48
2022-03-21T23:08:34 Starting Internal StellarSolver Sextractor with the 4-SmallScaleSolving profile . . .
2022-03-21T23:08:34 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
2022-03-21T23:08:33 Disabling the inParallel option.
2022-03-21T23:08:33 Not enough RAM is available on this system for loading the index files you have in parallel
2022-03-21T23:08:33 Evaluating Installed RAM for inParallel Option. Total Size of Index files: 0.303479 GB, Installed RAM: 0.83849 GB, Free RAM: 0.181896 GB
2022-03-21T23:08:33 Image received.
2022-03-21T23:08:12 Capturing image..</code>

This is with Use Scale and Use Position
drive.google.com/file/d/19boQuVxB9kV-v4A...a3/view?usp=drivesdk
3 months 1 week ago #81619

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

  • Posts: 2722
  • Thank you received: 760
Ok I just ran your first image, solver.fits in the StellarSolverTester program I made to test Solving and here are the results:

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Starting Internal StellarSolver Astrometry.net based Engine with the 4-SmallScaleSolving profile. . .
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Solve Log Odds: 99.4057
Number of Matches: 19
Solved with index: 4108
Field center: (RA,Dec) = (102.905, -15.207) deg.
Field center: (RA H:M:S, Dec D:M:S) = (06:51:37.103, -15:12:25.164).
Field size: 66.405 x 49.7563 arcminutes
Pixel Scale: 1.9644"
Field rotation angle: up is -13.1734 degrees E of N
Field parity: negative

Successfully solved with child solver: 6
Shutting down other child solvers
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Internal Extractor w/ StellarSolver took a total of: 0.526 second(s).
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

I also tried it with the LargeScaleSolving profile as you did, and it did not solve since the field size is 66.405 x 49.7563 arcminutes

And then I tried the second image you sent solver2.fits and here are the results:

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Starting Internal StellarSolver Astrometry.net based Engine with the 4-SmallScaleSolving profile. . .
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Solve Log Odds: 215.27
Number of Matches: 35
Solved with index: 4107
Field center: (RA,Dec) = (105.673, -8.49984) deg.
Field center: (RA H:M:S, Dec D:M:S) = (07:02:41.505, -08:29:59.426).
Field size: 66.5251 x 49.8034 arcminutes
Pixel Scale: 1.9671"
Field rotation angle: up is -13.2995 degrees E of N
Field parity: negative

Successfully solved with child solver: 6
Shutting down other child solvers
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Internal Extractor w/ StellarSolver took a total of: 0.516 second(s).
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

And just like the other one, I tried the LargeScaleSolving profile and it did not solve.

If you look in the parameters of the large scale solving profile, it makes the minimum width about 10 degrees, but your images are about 1 degree, kind of like the image scale on my telescope, which is interesting since I have twice the focal length you do, I think.

Note that they solved with 4107 and 4108, so you should make sure you have these index files (and the ones nearby them) installed.
3 months 1 week ago #81620

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

  • Posts: 41
  • Thank you received: 5
Hi,

I set it now to 4 - small scale, reset to default, and I am. getting mixed results, bit atleast I am getting some results. When I do not use "Use scale" I get better results, but I have most ofmindex files loaded, should I remove ones I do not use to increase chance to solve?
3 months 5 days ago #81667

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

  • Posts: 41
  • Thank you received: 5
So. i am. having mixed results with basically same image. first solve is successful and it is no scale but use position, as are next 4.

Then I turned on scale, then turn off position then turn off both, nothing helps. Would it help if I remove index files I do not use? I see your solver isnalways solved with "Solver 6" is that the thread based? I have rpi 4, so only 4 threads. Why always so high number?

ibb.co/4th74gm
ibb.co/J3cYgmp
3 months 4 days ago #81703

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

  • Posts: 2722
  • Thank you received: 760
It might make it solve faster if you remove index files that you don't need, but I haven't seen much of an issue on my system to have a decent sized set. I don't think having too many indexes has ever prevented a solve for me If it helps to turn off the scale, I would look carefully at your information in Ekos about your telescope and optics. KStars uses that information to inform the solver of the scale. If that information is incorrect, it will definitely prevent it from solving, so yes if you are sending it incorrect information, turning off that information helps. Every time StellarSolver solves, it can let you know the image scale so you can check it against the information you have. The same is true for position, if Ekos is way off in the position and you have "use Position" checked, it can definitely prevent a solve. The position and scale do not have to be perfect, of course, because that is why you have the solver to get that information. But giving a guess at the scale and position makes the solve faster and less processor intensive at the expense of possibly not solving at all.

Yes, it is definitely thread based, that is a key part of how the solver is so much faster. The number of threads is determined by the Ideal thread count QT determines. It is based on information about your system such as the number of processors and other info. I don't choose any particular number, QT does. The ideal thread count does not actually match the number of cores.
The following user(s) said Thank You: Outta
3 months 4 days ago #81714

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

  • Posts: 2722
  • Thank you received: 760
In looking at your images of your indexes, I think you should have some of the ones in the right column. I find that sometimes they solve when the ones in the left column can't solve it. In fact the images that you sent to me solved with the index files near the bottom of that right column if you will recall. I would definitely have those installed. You don't really need the ones near the top, unless you are doing really wide field stuff, but of course they take up very little room so it might be fine.
The following user(s) said Thank You: Outta
3 months 4 days ago #81715

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

  • Posts: 41
  • Thank you received: 5
Indeed adding this index files has solved most of the problems.

It solves 90% od the time now!
Thank you very much on your extensive help!

It is so good not to rely on having internet and slow nova.astrometry.net :)
3 months 1 day ago #81796

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

  • Posts: 2722
  • Thank you received: 760
sure, no problem!
The following user(s) said Thank You: Outta
2 months 4 weeks ago #81838

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

  • Posts: 41
  • Thank you received: 5
Hi, Me again

I have bought new scope, skymax 127 mak, and it is not solving once again. Everything is default but as i can see it is simply bot selecting real stars, and it is selecting random noise as stars... Any tips?

File Attachment:


nova.astrometry.net/status/5771052

Blind solve
Last edit: 1 month 1 week ago by Outta.
1 month 1 week ago #82894

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

  • Posts: 2722
  • Thank you received: 760
There are a couple of ways to deal with that, but whichever way you do it the key point is to make the star extraction ignore smaller objects and only extract bigger ones. You can do this by binning or downsampling the image for example. Or there are several settings in the star extraction settings such as minarea or minimum size. I would recommend taking a couple of your real images and playing with those settings to see what works well using load and slew in the daytime
The following user(s) said Thank You: Jasem Mutlaq, Outta
1 month 1 week ago #82895

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

Time to create page: 0.602 seconds