Dithering worked just fine in my test tonight with M17.

 

Thanks for checking!
That indeed looks like the guider cycle time is unchanged by the dither. Then again, it was already shorter than mine. Or it is somehow related to the camera? I use two ZWO cameras....
Unfortunately, weather prevents further tests here at the moment

I suspect this is an ASI driver/lib issue. Since Jasem doesn't use them, the issue doesn't show with his setup. Simulators guiding works as expected, too.



I might have a 20% chance to test guiding with my ASI combo tonight.

that would be a great test! I had, however, already tested the latest libASI, which did work with the old INDI, but not with the newer ones.

I managed to guide for a while although conditions were very bad. While guiding, I did not recognize any issues. However, I saw the Indi log window reported double requests for guide images after dithering. The amount of information in my logfile is overwhelming and I can't see whether it confirms the issue you reported. It's attached so everybody can have a look.

What I found though was the process of producing master dark images for the darks libary did not work at all. It crashed KStars multiple times when the master dark was computed and I never got a full run completed. After ~10 crashes I gave up.
 

Alfred,
this indeed matches my experience: There is a continuous stream of exposures from guiding at the requested exposure time exactly up to the point when, after dither, guiding is resumed. This resume triggers a new exposure, which aborts the currently ongoing one:and the guide module gets that aborted frame, which has only been exposed for a second or so, instead of the requested 4.5s. The time between readouts is basically given by readout time ad processing time. You can see that, e.g., the flux of the brightest star went down from 27902 before the dither to 9532 after dither.

As I mentioned in post #72049, that abort also happened in the 'old' version, but there only once, and after that only one stream of exposure/readouts continues with proper exposure times, whereas now both seem to continue.

Thinking about it, my guess is that the main error is that when an exposure is already going on at the time when a frame capture is requested, the ongoing exposure is aborted, but read out, and then returned as the requested frame, instead of completely dumping it and waiting until the newly triggered frame exposure ends.

I believe I found one issue that could contribute to this. Fix pushed to GIT: invent.kde.org/education/kstars/commit/7...a7c317b252fdec560fa9

Will compile & test ASAP!

I spent the night with the intention to test the latest guide module changes but ran into a wall of obstacles that prevented me from guiding. The two most important ones:

1. Star extraction/plate solving did not work as expected

No matter what parameters I entered, plate solving failed complaining there were too few stars in the field while in fact the frames were of good quality and did show many stars you could easily guide on. After a while of unsuccessful attempts to get it working, I changed guide/main camera and used the guide cam for solving. This time it did solve in a second! It did show the same FOV as the main cam before. The quality was not quite as perfect as with the main cam (still good though!) yet plate solving did work flawlessly and quickly. I then reverted to the main cam which again delivered an even better frame but Ekos still complained it was unable to solve.

I then shut down server and client, checked the mount, cams, etc. and started over. This time it was the other way round. The main cam did solve reliably and in no time but the guide cam did not solve, despite same settings, FOV was set correctly for both scopes.

2. Calibration was going nuts and I was unable to rectify the problem all night. It started out as usual, RA drifting forward nicely. The RA backward drift however was exactly in a 90 degrees angle against the forward drift. Of course I checked the usual suspects like RA axis latch, cables, etc. but everything was 100% OK. I tried over and over again but the result remained the same.

 

According to the logfile <:Mgw0500 and <:Mge0500 commands were sent. As you can see in the calibration plot the result was much different from what one would expect. Since I didn't get calibration done, I was unable to test the new guiding code. I'd be glad if someone with a better understanding of the calibration log had a look at the logfile. I'd really like to get a hint as to what went ẃrong since I am afraid I could run into exactly the same problem again. I'll try again tonight but am not too optimistic.

 

Alfred,

Sorry about all the issues. 

Re the plate solving: I see you are using ASTAP. I reported an issue a while back with stellarsolver/ekos & ASTAP where ASTAP expects FOV degrees for the image height, but is being sent the width degrees. This can be worked around by unsetting "Use Scale" or perhaps both unchecking "Use Scale" and "Use Position" in the align options  Scale & Position menu when using ASTAP. You can test on your own as well by using the offline astap program. For instance, on my system astap is installed as /opt/astap/astap. So you can run "/opt/astap/astap -help" to see its options, and, for instance, capture a fits you're having trouble with, and running "/opt/astap/astap -f FILE.fits" to solve it "blind" without scale nor position. You can see the args stellarsolver is sending astap from your log--where it's adding in the scale (that's the -fov 1.36103 in your log) and the position (-ra 19.1981 -spd 75.0158).

Re the calibration: Sorry, no idea. That's the first time I've seen something like that. I'll try and look a little more later.

Hy
 

Thank you so much for looking into it, Hy! That's good to know. Should I experience problems again tonight, I'll save a couple pictures that fail to solve so I can experiment and sort things out later. Looks like weather will be good so I'll definitely have another try tonight.

I am much better prepared now as I have several KStars/Ekos versions and also two indi servers ready for testing tonight. Hopefully I'll be able to actually guide.

Finally I got to guide a few  sequences so I'll leave the logfile for everybody to inspect. I got the solver working with ASTAP and astrometry as well. But calibration was a mess again. I'll write more about it later.