Live Stacking Ideas

[WB91]

I tried the autofocus routine last night and didn't like it at all. I don't understand why the settings are the way they are in sharpcap autofucus routine while some seems to be embedded within sharpcap so I could not change them. It was trying to do over 35 focusing steps which is overkill. This routine should really be mimicked from NINA, so easy that way. The only settings you really need to enter should be how many focus increments you want to make, so in Nina if you say 5, it will do five on either side of the bottom of the focus curve, so even if you are out of focus a lot, if you have not reached the bottom of the focus curve within 5 increments, it will continue until it does and then after the curve starts going back up, it will count 5 increments and finish, then automatically sends the focuser to the best position. The other setting you need to have is how many focuser motor steps you want to make per focus increment. So on my Celestron focuser I use a value of 30 steps, thus when the initial sequence starts, it will move 5 X 30 focus steps OUT thus 150 focus motor steps total, then focuses inward each increment consisting of 30 focus motor step moves which is 5 increments to bottom and 5 increments after bottom. Another setting that needs to be added is how long you want to set at each focus increment. With sharpcap because it does not fire a acquisition based on a triggers, it more or less fires continuously in the background, we need a way to tell the software to stay at a position for an amount of time before moving to the next i.e. if the imaging shots are running at 2s and we delay the movements to 6s, we should get 3 shots per increment which we could use an average or best score within the focus feature alone. It would be nice if we could just fire a an image based on the auto focuser routine itself but I guess Robin would have to let us know if it is possible as this is how Nina does theirs (exposure time and bin settings within the auto focuser settings). Regarding last nights testing, I got aggravated with it and I ended up writing my own sequence to get me a good focus curve, but when it was done, there is no command within the sequencer commands to send the focuser to best position. I also didn't see a command to open the multi-star feature either and then run the rest of the sequence, so the process was so long I ended up running Nina for autofocus alone and then go back to Sharpcap for acquisitions, which sucked because I had to warm and cool the camera each time.

Robin, if if it is too much of a hassle to make your routine work more like NINAs, could it be that a temporary fix could be added to the software to be able to open multistar focus lunch within a command in the sequencer, then we program our sequence, then we need a new command to send focuser to best position? If so, I can test this again and provide the sequence to people on this forum. Please let me know. Thanks!

[admin]

Hi,

a bit confused here, since a lot of the things that you say you want to change, you can change. For instance you can specify the distance that each step should move, the number of steps and the number of frames to capture at each step - see https://docs.sharpcap.co.uk/4.0/#Automa ... 20Scanning

The automatic refocus option (I *think* that this is what you are talking about) uses the settings for those parameters that were used the last time that you performed a successful assisted focus action - basically it is helping you by remembering the settings that worked for you last time and using those. If you want to override those settings and specify them manually in the sequencer, you can do that too - see the two autofocus steps listed here - https://docs.sharpcap.co.uk/4.0/#Focuser. In these, you get to specify the range of focus travel to cover and the number of steps to divide it into. You can't currently specify the number of exposures at each step, I think it uses the settings that you used manually the last time, but that could be added.

I must admit that I have considered the idea of stopping any scan early once a good result is found and there are a good number of data points on each side of the best focus point - I agree that is something for the to-do list.

cheers,

Robin

[WB91]

Robin, have you run the NINA autofocus routine while using overshoot, or the same using the ASIAir so that you can streamline the autofocus routine in sharpcap i.e. a one click operation as they do? Have you done any videos on how to set this up because I still find this a nightmare to deal with. If not, I would like to create one for ya so if you wanted to communicate with me direct so I can properly make one without discrepancies, that would be great.

Also, following up on the previous messages, any improvements to live stacking? or special sequencer commands to be able to move to the best focuser position, etc?

[admin]

Hi,

my take on the biggest difference between NINA and SharpCap is that NINA looks a bit smarter on working out the range of focus values to scan over for itself - for instance scanning a bit further if it has not got enough points on either side of the possible best focus and stopping short if it already has enough.

Additionally, NINA is set up on the assumption that you are starting near best focus, whereas the manual SharpCap scan is based on the idea that you are starting on one side of best focus to run past best focus to the other side. That could do with fixing.

The refocus button (shown below) is based around starting at/near the point of best focus and will run a fully automated focus around that point. The settings used will be based on the most recent successful focus scan that you have run. This can also be activated from the sequencer.

Capture.JPG (39.72 KiB) Viewed 1611 times

cheers,

Robin

[WB91]

Robin,

I have been working on a video on this focusing method and almost done, but one thing I have to do is manually offset the focus start position in order to get a full curve. Example, say I move the focuser to a place close to focus, if I set my focus steps at say 5 with a total of 16 positions, it will only move out 16 position which means I may not see the full down and up curve. Could you add an option to split the difference, meaning, if the position movements is every 16, the focuser would move inward 8 positions, then start moving out 16 so that we generate the complete curve? This would be perfect for what we are trying to achieve. I hope to shoot again early friday and saturday morning so let me know if this is something you could add and send me to test and I would gladly put out a video on this and other Sharpcap features

[admin]

Hi,

that's exactly what the refocus button should do - it looks at the total movement of the last scan (maybe 16 measurement points that are 10 steps apart for a total of 160 steps). It will then move 80 inwards from the current position and scan outwards from there for a total of 160. If it is not doing that then please let me know, as it is supposed to!.

cheers,

Robin

[WB91]

Robin,

I was able to do more testing on this feature last night and once again I want to say thanks for all your hard effort to get this to work in SharpCap!

I would like to report a few things and share some video but I dont want to post it on this forum yet until I can get some clarification from you first, as I do not want to put something out that might not be correct or misleading. Please shoot me an email so I can share with you my finding and then I can put out a very detailed online instruction video to get all these features you have in something more structured and to better help you promote your software. email: b********@*****.com

Regards,
Bill