Page 1 of 1

yet another noob :)

Posted: Wed May 17, 2017 8:07 pm
by mstiles
would it be possible for someone to look at my data and see what i'm NOT doing, or, maybe DOING to get consistently yucky results?

this is the unadjusted output of dss, 32bit..

https://www.dropbox.com/s/w4xwn3risfnn9 ... s.TIF?dl=0

here is my best attempt at it so far.. i think i cheated and only used Develop after the first autodev/bin/crop/wipe...

https://www.dropbox.com/s/dof749lefm2kd ... S.TIF?dl=0


when i don't "cheat" i end up with the below screenshot in the denoise module.. yuck

https://www.dropbox.com/s/yrtoo3rb1mipb ... 6.png?dl=0


never used the dropbox before, so i hope i did all that right ( first time for everything )

Re: yet another noob :)

Posted: Thu May 18, 2017 6:59 am
by Bill55
Who said it was easy? Don't give up, I'm still learning as well (after nearly 3 yrs). There are some brilliant guides in the User Notes forum section written by a guy called Guy. Work through them and take some time to experiment with the settings. Its a steep learning curve.
I'm sure someone on here can take your image and do something brilliant in a couple of minutes. For me it took too long to open your dss image so I gave up. The other images look like something good should come out.

Re: yet another noob :)

Posted: Thu May 18, 2017 2:39 pm
by mstiles
thank you Bill!! i'll take a gander at those notes

Re: yet another noob :)

Posted: Fri May 19, 2017 1:53 pm
by admin
Hi there!

I had a quick look at the data set.

The data set appears to have been color balanced somehow (and possibly pre-processed by some other means). Can you confirm you are using the settings as seen here?
Quick Start Tutorial: a quick generic work flow

Image
^ Giving StarTools virgin data is of the utmost importance. For example, if you are using DeepSkyStacker, make sure 'RGB Channels Background Calibration' and 'Per Channel Background Calibration' are set to 'No'.
Can you tell us a bit more about the acquisition details, and you gear?

When you are ready to do your "final" stretch with AutoDev (where you say you "cheated" by using a more manual Develop), be sure to specify (e.g. click and drag) a Region Of Interest over the object (or a slice of it). Without doing this, AutoDev will optimise dynamic range for the whole image rather than just the small object of interest. The latter is great for screen-filling nebulosity, but needs a bit of help in case the image is only partially "interesting" as is the case in a wide filed like this.

Does this help?

Re: yet another noob :)

Posted: Fri May 19, 2017 11:38 pm
by mstiles
Thank you for the reply Ivo... thanks to Bill's tip about the User Notes, i've been making PROGRESS!!

i now have the dss settings set to No on both background calibs... also am outputting the 32 bit fits ( 16 before )

i've spent all day reprocessing 3 of my other images and have made each one of them vastly better ( they still probably suck ) than before.. again, thanks to the user notes ( a MUST read(s) if you are like me and don't know *h** from apple butter about all this stuff )

i'm re-stacking the M57 data now.. oh, acquisition... using an ED80T-CF, canon t6i (unmodded).. i took 120 30sec subs at iso800, 30 darks and flats, and re-used bias frames from 2 days earlier

my guide scope and camera is in-route ( yay, something else new to learn ) and i haven't gotten a field flattener yet, but desperately need one.. i have a feeling i may need more data for M57 to shine, but i'm going to give it another try using the new "stuff" i've learned..

i did get over-zealous and posted a couple on astrobin already :roll:

http://www.astrobin.com/295889/C/

and

http://www.astrobin.com/296141/?nc=user


i hope those links work

Re: yet another noob :)

Posted: Sat May 20, 2017 6:44 am
by admin
Ha!

That's pretty awesome stuff you're producing - very good really for someone just getting started.

The M81/M82 image could - if fidelity allows - possibly benefit from some deconvolution (there's some really cool/unusal reddish outflow detail lurking in M82). You can increase fidelity a little by Binning (a little). It's a tradeoff.

Are you using a light pollution filter at all? How do you calibrate your colors in StarTools? Is this the default it comes up with?

Re: yet another noob :)

Posted: Sun May 21, 2017 2:41 pm
by mstiles
thanks!

beginner's luck obviously

no filters, and (from the User Notes) used a star mask and Sample to get the initial color.. then did the adjustments like the attached screenshot

i've been trying the Bin, but i don't think i know how to bin "a little".. i seem to loose all of the colorful "fuzz" afterwards.. can you create a mask in Color to just target the fuzz you want to mess with?

guess i should play more

Re: yet another noob :)

Posted: Sat May 27, 2017 8:33 am
by admin
mstiles wrote:thanks!

beginner's luck obviously

no filters, and (from the User Notes) used a star mask and Sample to get the initial color.. then did the adjustments like the attached screenshot
Colors look good - yellowish core (older stars), bluish arms (younger stars). I'd expect some red in the "cigar" though...
i've been trying the Bin, but i don't think i know how to bin "a little"..
Normally, in other applications, binning will only let you integrate an integer number of samples (e.g. 2x2, 3x3, etc.). In StarTools you can choose any value you like by setting the slider. Binning "a little" in this case is reducing the resolution just enough for a little fuziness of the detail.
i seem to loose all of the colorful "fuzz" afterwards.. can you create a mask in Color to just target the fuzz you want to mess with?
You can, but I'm not sure what you mean by losing the colorful fuzz?

Re: yet another noob :)

Posted: Wed May 31, 2017 12:58 am
by mstiles
Thank you Ivo.. I will practice more with binning

Currently attempting longer exposures while tackling phd2...

Liking M104.. only ONE thing to focus on. :)

I think the "fuzz" I was talking a about was the ends of the galaxy's arm(s)