Useful answers from Dirk to some questions raised on Facebook:

Exposure time to avoid saturation on single stars?
There is no need to avoid saturation of all stars in the image. That will be next to impossible in some fields. For most of our data the bright end saturation sets in between 9th and 10th magnitude. So, choose whatever single exposure time you use for your own purpose (or what is most efficient) and do repeat observations and stack those. The photometry calibration we do will identify the saturated stars and they are removed from the analysis. So there is no problem with this.

To dither or not to dither?
I can’t see any reasons to do it or not do it for our purpose. If you have good calibration frames (dark/bias and flatfields) then any variation in sensitivity of the different pixels are taken care of. Again, please do what you need to do for your own purpose. After submitting a first few test frames we can evaluate the quality of your data and suggest improvements if needed.

Rejection algorithms during stacking?
This can or can not be a good idea depending on your data. Usually a simple median stack works well as it removes any cosmic ray hits, as well as satellite or airplane trails. However, using a rejection algorithm (sigma clipping) with averaging usually gives a slightly better signal to noise ratio in the resulting stack. The differences however are small. The clipping can cause problems with removing stellar flux if you have quite variable seeing or under-sample the stellar psf – i.e. if your pixels are larger than your seeing. So again, it depends, but generally both will work just fine for us with minimal differences in data accuracy.