Latest autofocus routine is broken?

Discussions, Bug Reports and Issues related to Beta versions of SharpCap
Post Reply
opestovsky
Posts: 9
Joined: Mon Mar 22, 2021 11:25 pm

Latest autofocus routine is broken?

#1

Post by opestovsky »

I don't remember the version number, but I downloaded the latest beta last night at 10 pm CDT.
I ran an autofocus routine and clicked "best position", and Sharpcap crashed to desktop. It gave three options to report the crash, and I chose the first "web" option. But that seemingly didn't do anything, so I'm posting here.

I then restarted Sharpcap, and ran the routine again. This time, the routine couldn't complete because the camera was stuck in a loop - it would count up 0.3 seconds in the progress bar of a 2 second exposure, and it would loop back to zero, repeatedly.
User avatar
admin
Site Admin
Posts: 13122
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Latest autofocus routine is broken?

#2

Post by admin »

Hi,

I did have an autofocus related bug report yesterday - from the details it looked like the camera had disconnected from the PC while the focusing was happening (this was someone using an ASI533MC - not sure if that was you).

Certainly the fact that you had problems getting the camera to work the second time would indicate a more basic problem than the focus calculations - if that problem persists please capture a log which I will investigate. If you see the autofocus crash again, please try to submit the bug report but also grab a screenshot showing the bug report window - that will let me work out if I am seeing your reports.

thanks,

Robin
rac19
Posts: 142
Joined: Wed Apr 21, 2021 2:01 pm

Re: Latest autofocus routine is broken?

#3

Post by rac19 »

I downloaded the very latest Sharpcap 4.0 beta version tonight and single star FWHM focus assist just didn't work. FWHM was high when in focus and got lower as it went (visibly) out of focus. Everything was normal in Sharpcap 3.2, Thank goodness both versions can coexist :).
User avatar
admin
Site Admin
Posts: 13122
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Latest autofocus routine is broken?

#4

Post by admin »

Hi,

thanks for the report - that's odd since the single star FWHM is hardly changed since 3.2 (multi star changed lots). I just tested in simulation and it seems to detect blurred stars as having higher FWHM as expected.

Please can you capture and share with me 1 in focus frame, 1 out of focus frame and a screenshot showing the settings for the FWHM tool and your selection area around the star you are using - from that I should be able to recreate what is happening and hopefully understand the problem.

thanks,

Robin
rac19
Posts: 142
Joined: Wed Apr 21, 2021 2:01 pm

Re: Latest autofocus routine is broken?

#5

Post by rac19 »

I found it all quite baffling too. The U curve seemed to be significantly offset (outwards) in defiance of the star image. I wonder whether or not I had the exposure completely wrong in 4.0 and nearer the mark in 3.2. I will be trying again tonight but at f/2 not f/5 as it was last night. I will at least capture done screen shots if I have a problem.

On the subject of exposure (for the focus assistant), it would be useful if a simple FWHM bell curve was available so that exposure could be adjusted to give a nice peak.
rac19
Posts: 142
Joined: Wed Apr 21, 2021 2:01 pm

Re: Latest autofocus routine is broken?

#6

Post by rac19 »

I used the single star FWHM Focus Assistant tonight with Hyperstar (f/2) with success. This might by normal but I got different results with a bright star (Arcturus) than with a dim star nearby. The lowest FWHM for the dim star was at a focus position possibly 1/4 turn further out than for Arcturus.The best FWHM was also lower for the dim star which I suppose is to be expected. I have used the dim star focus position to capture images. It looks quite well focussed during capture. I will see how the post processing looks tomorrow.
User avatar
admin
Site Admin
Posts: 13122
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Latest autofocus routine is broken?

#7

Post by admin »

Hi,

the FWHM is going to be more reliable on a dimmer star (one that is not saturated), since with a fully saturated star it is impossible to tell what the maximum brightness should be.

The multi-star FWHM option automatically looks for stars that are not fully saturated and measures those - that is my focus tool of choice these days, so the single star FWHM has not received so much love and attention recently. If you want to see how bright your target star is, you can use the histogram along with the selection area tool to see the pixel values, or use the new pixel value readout tool in SC 4.0

cheers,

Robin
Post Reply