Optimum work flow from Sharpcap?

[timh]

Hi folks,

This is the first question that I have posted. The forum is excellent and Robin's answers have usually been so clear that I have found that most questions that I have had in the past have already been answered!

My question is this. Ideally I like to combine EEA Astronomy with the process of capturing high quality data to then work with off-line in SIRIL or DSS and then PixInsight in order to produce good images. So I do like to be live stacking if at all possible. However most of what have I read suggests separately capturing dark frames, bias frames, flats and light frames and then working them up outside of Sharpcap. Nevertheless I notice that Sharpcap has many useful features in Livestack -including the ability to subtract the darks and flats on the fly, filter by FWHM and brightness to avoid bad frames and then also the ability to automatically collect a long series of say 4 or 6 minutes stacks which can be saved as 16 or 32 bit FIT files. These I assume are already corrected for Darks and Flats?

So can I have my cake and eat it? Enjoy the EEA aspect in Livestack while just relying on the timed Sharpcap autostack facility to create short FIT file stacks, stack these off line (without having to bother with the darks, flats etc) and end up with images of comparable quality to comparable quality to those I would have created via any other workflow?

best wishes
Tim

[admin]

Hi Tim,

some people do just this - they set the live stack to auto save and reset every 10 minutes maybe and then collect 12 stack images over the course of two hours and then bring those together in some other processing application. If you are using darks and flats in SharpCap then those images contain the corrections for the dark and flat frames, meaning that you don't need to worry about those in your later stacking.

On the other hand, if you are using the option to save the raw frames while you are live stacking, those raw frame images are uncorrected. This gives you the chance to either reprocess the data entirely outside of SharpCap or work with the live stacking results depending on your preference.

Cheers, Robin

[timh]

Ok many thanks. I'll have a go with raw frames too at the next opportunity and see if there is any real difference (i.e. maybe raw frames that should have been rejected but that were not caught be the FWHM and brightness filters) versus just using the 'preassembled and quality filtered' 10 minute stacks. Tim

[timh]

Hi Robin,

I have followed through on this and, having first sorted out how to correctly apply flats and darks, it all seems to work nicely with the DSS stack of raw files plus the master dark and master light looking indistinguishable from the DSS stack obtained instead by combining just the cognate Sharpcap debayered autostacks.

However. I am not sure what if anything can be done with the 32bit 137Mb debayered stack FIT files? The 16 bit files seem to work fine in DSS and also SIRIL another stacking program I have tried. As far as I can tell (perhaps someone will correct me?), PixInsight seems to work only with the raw files plus lights and darks etc. and won't stack the Sharpcap autostacks.

So my further questions are two fold.

Firstly do you know of any software that will stack the 32 bit stacks that Sharpcap produces ?

Secondly under what circumstances is it even likely to ever matter whether I use 16 or 32 bit files? Typically I am producing 10 minute stacks of 15-20 subframes under skies that are 'OK' at best. With low gain I think that the theoretical best bit depth for a raw frame on my CMOS OSC is about 12? Is it possible to do some sort of calculation of signal to noise that might say, for example, that a 16 bit stack will always be sufficient to adequately capture the information from say X (4 ?) subframes and, in practice and under 'normal' seeing conditions up to (say?) 16 or whatever?

So here I am just wondering whether 32 bit stacks become more desirable the more subframes they comprise and the better the seeing conditions? Or am I thinking of this incorrectly?

best wishes
Tim

[admin]

Hi Tim,

I must admit that I don't use the 32 bit saved images myself – that feature got added because somebody asked for it.

If your camera is producing 12 bit images then you can add 16 of them to the stack before you have any risk of information loss in a 16 bit image. Beyond that point you have to start thinking about the amount of noise in the lowest order bits of your 12 bit images – for instance if the noise is for ADU at minimum then that will be occupying the lowest two bits of the 12 bit data meaning that you could extend the individual stacks from 16 to 64 frames without loss of data in that case.

Cheers, Robin

[timh]

Many thanks Robin,

Since I am only stacking about 16 frames I will stick with what I am doing then and that seems to work.

Incidentally I have just answered my first question if it is of interest - PixInsight works fine to stack the 32 bit files from Sharpcap in fact. First put them into the Registration tool (with one as the reference) and then do image integration on the registered files.

best wishes
Tim

[timh]

Actually -- I've just realised a) that my AS1294 camera is 14bit rather than 12bit ADC and b) that I haven't been using the Brain properly and that for maximum dynamic range (what is unity gain for?) I should probably be doing many more shorter exposures under the rather bright (5e or more) skies? So maybe the 16 / 32 bit case is a bit more marginal (but I suppose that if skies are brighter it is also the case that more of the signal is noise?).