Oknoder Loc: Western North Dakota

First thing I did in DSS is to add all my calibration images. My darks and bias were shot months ago, since the dark current and sensor noise really only changes based on outside temperature. It is advised to redo these images every year or so. The flats should be taken along with your light images. The flats will help eliminate all the defects and impurities in your image train, mainly vignetting and dust which show up as donuts in images.

After loading my lights, I click the check all option then choose to save the file list, just incase the program crashes. I do not understand how DSS scores images so I did not play with the option check above a threshold.

In the option register checked pictures, I select the three main options. I choose to select the best 70% for the stacked image, which in this case is only 10 images. Most of the images I included were 10min subs at ISO-1600, I did include the longer exposure times also to help increase my signal to noise ratio, which DSS lists as SNR. Under the Advanced Tab I select the reduce noise by using a median filter. I run the star detection threshold which will help decide which alignment algorithm I will be using.

Within the Recommended Settings it advises me to use AHD debayering, no idea what AHD means but I will follow its recommendations. I am not using narrowband images so I leave super-pixel option unselected. I select setting the black point to 0, and will change this after stacking. I also use, per channel background calibration.

Under stacking parameters;
Result tab;
I chose standard mode since the only changes from image to image is a very minor dither of 5-10 pixels, I have never been able to get DSS's drizzle options to work without the program crashing due to a memory error, which I think is due to the large file size and the program being only 32bit so only 4gigs of RAM is available to be used. I do select the option Align RGB Channels in final image.

Light Tab;
Because I am unfamiliar with DSS I really do not know the best option here so I selected Median, and hot pixel removal. I do not know what dark optimization and dark multiplication factor are though so I did not select them.

Flat and Bias/offset Tabs;
I chose Median

Alignment;
Bicubic is the method I will be using since during my threshold detection the program detected around 200 stars to use in the registering and stacking.

Intermediate Files;
I am not really sure what the first two options dealing with creating calibrated images of each frame is for, so I skipped these options. I do always chose FITS format since usually you can fit more data in an FITS file. I chose this here without knowing the type of FITS DSS uses but an unsigned floating FITS file can hold 32bits of data whereas a TIFF is restricted to 16bits.

Cosmetic;
Under both option of removing Hot and Cold pixels I chose the same parameters of 3 pixel filter size with 33% detection threshold. I chose to replace the rejected pixels with a value of the gaussian filter.

Output;
Options are pretty much self expainatory

I change the temp folder location to my scratch drive to help reduce processing time. Since I am using fairly quick processors I do not feel the need to reduce the worker thread priority, but those with slower machines will probably benefit from selecting this option. I use all available processors, even though on checking DSS only uses half of the available cores.

When DSS has finished aligning and stacking it shows a represented image of the data. There is already an image that DSS autosaves automatically so I feel free to take the liberty to tweak the image to try and get the most out of the data acquired. I adjust all the color channels individually until they are all in the same place of the histogram. I then select the option to link settings, and then try to stretch the histogram as much as I can. Since each image will have different parameters there really is no viable advice to give on this, other than knowing that this data will be scaled from a 32bit file into a 16bit container so if nothing is done here half of the data can easily be lost and the image will become excessively dark, because the majority of your data is on the dark side of the scale. Best advise is to make small changes and click apply often to see if your adjustment are moving the image in the direction you are visualizing in your head.

The attached image is straight out of DSS, normally I would take this TIF into photoshop and play with it to try and squeeze out as much data as possible, which I may do, but I will probably go through my normal routine, since I do not have a firm grasp on how to exploit all I am sure DSS has to offer. From what I can tell DSS fell a bit short of my expectations only along the lines of the integration of the Flats, since I see remnants of the dust donuts and vignetting.

(Download)

Albuqshutterbug Loc: Albuquerque NM

Aside from the stacking noise this is a really nice stack.
I think I mentioned somewhere that I was starting to dislike you just a little bit.. ;)
All kidding aside, a little bit of post prod will make this a very, very nice stack.
Mine never come up this clean even when I get it to work. Just look at my beehive shot.
I played with this a bit in lightroom.
May I post what I did and tell me how far off I am on what it should be?

JimH123 Loc: Morgan Hill, CA

Thank you very much for the stacking instructions. I am still getting the feel for these settings. I have now added dark frames and I will work with one new thing at a time. I am finding my result is dimmer that the originals I started with and I have to boost it so much in PS. Must be a setting I have set.

This is a fabulous image you posted. Once it is further processed and the noise removed, it will be even better.

CraigFair Loc: Santa Maria, CA.

Very nicely done Matthew. Love your in-depth information.
Craig

Oknoder Loc: Western North Dakota

As for the dim results before saving your image after stacking with DSS it is best to adjust the histogram by playing with the color sliders.

On a side note this image is the result of my normal stacking software which does a much better result in applying the darks,bias and flats to minimize the noise.

(Download)

CraigFair Loc: Santa Maria, CA.

Again nicely done Matthew. Love you work and critiques.
Craig

Oknoder Loc: Western North Dakota

Thanks Craig

Albuqshutterbug Loc: Albuquerque NM

The small circles within the stars are courtesy of DSS or post prod? On zoom it's on all of the stars and muddies the shot. I liked the other post better.
Sorry.
Jim

Oknoder Loc: Western North Dakota

I think what it is, is an artifact of the deconvolution and the compression down to an 8bit file for posting on this site.
I looked at the original master file and it is not present but it is a common artifact of deconvolution, which is a sharpening option in CCDStack.

No need to be sorry I did not really look at the jpg before posting, which I should know better. This is the exact type of constructive criticism that helps us see our errors and pushes us into producing better images.

Thanks Jim, I will see if I can compress the image and leave the artifact behind. Good catch.
Matthew

Albuqshutterbug Loc: Albuquerque NM

If my shots only looked this good.
You really did see my attempt right?
Even with artifact you blow me away.
;)

Oknoder Loc: Western North Dakota

How odd I thought your captures were better with more definition. I told myself it was because you are using an 8" whereas I only have 6". Lol as long as we keep getting better we all win.

Thanks,
Matthew

JimH123 Loc: Morgan Hill, CA

The AT6RC uses superb mirrors and is quite capable of very good images. And then with the auto-guider, its hard to compete!

I have an 8", but no auto-guider. I won't get the same pin points no matter how hard I try.

But I did talk to a SW company that is considering (read "considering") to provide SW that is for astrophotography and that it would figure out the tracking movement and to pull those movement back into a single point also adding the accumulated light to that point.

If this SW happens, it would allow anyone with a drive to capture longer and to process it back into an image without trails. It would also have an advantage with noise. If it doesn't have a trail, it is noise and can be dealt with.

Would probably help those without drives too by allowing them to shoot for some to-be-determined amount of time based upon FL and to retrofit it into an image without trails.

I don't know if or when this SW is really going to happen, but it intrigued me to consider it. And why not do this? We stretch images to bring out fuzzy galaxies, so why not compress trails to undo the rotation of the earth?

astroturf Loc: vacaville ca.

Nice round stars with 10 min. subs and fine post processing. Excellent