SharpCap Forums

While using Version 3.1.5059 tonight I believe that my flat frame was not being applied to Live Stack. However, the flat frame appeared to be correctly applied during a live view.

I determined this because there was a rather prominent dust donut on the lower left. In live if I selected and deselected the flat frame the donut would disappear and reappear. But during the Live Stack the donut was there no matter what I did with the flat frame.

Yes, that looks like it's a bug - I will investigate and fix.

thanks for the report.

Robin

I tested back to version 3.1.5027 and found the problem was present even then.

Hi,

yes, having a look at the code I think the case where the camera is taking RAW images and you are using flats is not handled in live stacking. MONO or RGB modes from the camera will be fine I think.

Can see a couple of ways of fixing, but not sure which is the best yet.

cheers,

Robin

Yes, MONO and RGB both have the flat applied.

Thank you for the work around. Until the fix is available I will use RGB for EAA.

Hi,

please have a look at this build

https://d.sharpcap.co.uk/download.html? ... 3.1.5076.0

It should solve your issue with flat frames. I have had to make bigger changes to the code than I would like at this stage of a release, so while I have tested the new code myself, I'd advise testing on terrestrial images if you can before testing on a clear night - you can live stack non astro images if you turn off the alignment.

cheers,

Robin

I still have issues with live stack and flat frames also with latest build 3.1.5092, and QHY163 camera.
With my altaz mount I have large field rotation which is well compensated during the live stack process; unfortunately, also the flat frame is rotating, as visible in the attached foto, mode 16 bit bin 2x2.
the result is an incorrect flat frame application with residual gradient and dust visible in the stack.

the problem is not there with normal video capture, but only livestack.
Can you help?

Hi,

can you share a raw frame of the target and the flat frame for me to test with - also a dark frame if you were using darks. The files will be too big to upload here, so probably best to upload to somewhere like dropbox and then send me the download link.

thanks,

Robin

Hi,

I can't make this happen in my testing, but I do have a question - were you using sigma clipping in the stacking?

The reason for the question is this - it looks like you may have a flat that is overcorrecting giving bright spots to the top right of the image. If that's happening then those bright areas will become the accepted mean value for the stack in that region. As the field rotates other areas of the frame which are not over-corrected by the flat will come into that area but will be rejected as being too different from the existing values.

If you were using Sigma clipping then maybe have a go with default stacking to see if that helps.

cheers,

Robin