I'm not very happy with this image and am wondering whether it is because of the original data quality or my lack of experience with StarTools. The image was taken with a Canon 7D at prime focus on a Celestron 8" EdgeHD with the 0.7 focal reducer. Its a DSS stack of 30x120 sec lights, 30 x 120 sec darks and 30 bias frames. No flats.
I'm hoping for some guidance on whether to concentrate on getting more data and adding flats or whether I need to concentrate more on learning StarTools. In StarTools I followed the sequence in Marco's workflow log.
I tried uploading the .tiff image but the upload failed (the file is 11MB) so I uploaded a .jpg instead.
Pinwheel Galaxy.jpg (452.29 KiB) Viewed 4441 times
I am a relative beginner at StarTools too, and hopefully some of the more knowledgeable posters will be along soon, but in the meantime, how are you focussing? Some of the stars look bigger than I would expect, (may be nothing to do with focus, so don't take this as criticism).
Skywatcher 190MN, ASI 2600 or astro modded Canon 700d, guided by OAG, ASI120, PHD2
Difficult to draw any conclusions without access to the raw data but from what I can see the jpg seems to profoundly clipped to black (could be this monitor I'm using but the background looks flat black). So the faint structures are completely lost. Perhaps that was the result of an attempt to suppress background noise? The somewhat bloated appearance of the stars may be the result of poor focus, or the result of processing without using a star mask, or indeed - that's just the way an SCT displays stars. Again, difficult to say without seeing the raw data.
More data is always a good thing, and not just more subs but longer ones (I routinely used 300s with my 70D and have pushed it out to 900s @ 200iso on moonless nights). Using a large dither with a DSLR is a necessity really for reducing background noise (colour mottling). Flats are useful if you are getting gradients or vignetting, Darks less useful unless they are truely representative of non-random noise (something very difficult to achieve in the absence of set-point temperature control - which is what you have with a DSLR). Your darks may end up adding noise rather than cancelling it out. I would rather work with 60 light frames than 30 lights + 30 darks in this case (and even better to have 20 lights @ 360s each in this example!).
Thanks Chris, you have given me a lot to think about, both with gathering data and then processing it. I did push the grain size up to produce a smoother looking background and didn't use a star mask so that maybe part of the problem.
I must admit that I struggle a bit with StarTools.
I haven't really a clue what the notes at the beginning of denoise mean.
The stuff about noise floor and grain size don't really make any sense to me. I don't know how to distinguish read noise from any other noise.
On some of the other functions changing parameters doesn't seem to make any difference so I don't know if they will have any impact further down the line.
Other than following Marco's log I'm not sure which functions I should be using and in what order.
Decon is a mystery to me.
My only experience with this sort of thing is a little bit of Photoshop and a lot of Lightroom. In fact putting that image in Lightroom and increasing the saturation and reducing the noise produced a better result. But I believe I should be able to do it all in StarTools.
Ah well, onwards and upwards.
Thanks again,
Best wishes,
Robin