Plate Solve Error?

[rac19]

I had a problem with Plate Solve and Sync last night. It seemed like the correction (say 0.10 degrees) was in the wrong direction (RA). Such successive Plate z solves seemed to step the same amount in the wrong direction each time. I tried horizontal image flip, which I don't think is normally necessary, but that simply reversed the problem. GoTo was spot on after each Plate Solve and Synch. I really did seem as though the direction of RA correction was reversed. I am using the latest beta version which expires at the end of June.

[Tiago Ferreira]

Just want to add here that since the last update i couldn't plate solve anymore. Always gives an error. I used 2 days ago but due to bad weather, i couldn't use it anymore.

[admin]

Hi,

were you using Astap? There is an issue in the current beta version with Astap plate solving being wrong. I have a fix for today's update.

cheers,

Robin

[Tiago Ferreira]

It said ANSVR not solved...... and something you ask about changing number of stars.
But i didn't change anything.
The error was too fast and i couldn't read it all.
I will test again soon because forecast is better now.
I also tested with the old method, the virtual telescope and happened the same so it should not an issue with the new push-to...

[admin]

Hi,

Ok, so if it is Ansvr and the message is about stars then it is likely that there are either too many or too few stars being detected. The number detected depends on the 'sigma' setting in the SharpCap plate solving options and also the nature of the image (noise levels, brightness, etc). The best thing to do is to look at the log where you will find the message recorded so you can read it properly.

cheers,

Robin

[rac19]

I have just shut down due to approaching rain but I am pretty sure that I am using ANSVR directly. I don't get any error messages, it just seemed that the RA connection was in the wrong direction. The 'scope connection was via CPWI and it has occurred to me that I have GoTo Approach set to Reversed in CPWI. I am sure that I have set it this way before without it affecting Plate Solve correction. I haven't looked in the log. It looks like clear skies tomorrow night so I should have a chance to try again then.

[Tiago Ferreira]

Hi,

Ok, so if it is Ansvr and the message is about stars then it is likely that there are either too many or too few stars being detected. The number detected depends on the 'sigma' setting in the SharpCap plate solving options and also the nature of the image (noise levels, brightness, etc). The best thing to do is to look at the log where you will find the message recorded so you can read it properly.

cheers,

Robin

It could be a coincidence to happen after download the last one. Although i didn't touch the sigma and always worked for me at 200.
The region (eagle nebula) may not helped but in the past it worked. I suspect the cloudy weather had to do with it. I will confirm soon.
Robin, does the stretch strength applied have influence in the software behavior? Might stretching less "help" the plate solving or doesn't matter at all?

[admin]

Hi,

sigma=200 sounds really high to me - I tend to use values in the 30-50 range.

As for the display stretch in SharpCap, it has no direct effect on the plate solving (the image is sent to Astap/ANSVR without the stretch applied). However, it has an indirect effect... If you have a stretch applied then you might tend to stick with lower gain/exposure values because the stretch is bringing out detail that would otherwise be too faint to see. That might mean that you are sending very faint images to the plate solving application, which combined with a high sigma might mean it fails to find enough (or any!?) stars.

cheers,

Robin

[Tiago Ferreira]

The gain i was using was the maximum, 550 in the zwo 294. This is because i search objects at high gain to see them better before live stack.
So it's a mystery to work 1 full year with sigma at 200 successfully.
Back then i troubleshoot here this problem and i think the default sigma was 50 and not working for me. After playing with it i got the 200 to work and never touched it again.
I did this without even knowing what sigma means. Just trial and error.

[Tiago Ferreira]

I just tried again and it worked fine! I also tested it again in the same Eagle Nebula and worked. The only moments it failed was because of the strong wind but it immediately recovered and solved.
It still keep solving even if we close the push-to box. We have to close and then cancel the plate solve for it to stop.
I didn't change the sigma, anything.
Robin, it really would be helpful a collapsible box so we can have only on the screen the important things which are the direction to go, while we are searching from a distance. Otherwise it covers a big part of the screen hiding the sky.
I'm really happy with it!