Newbie having trouble with, well, everything...

[JamieInCLT]

Hi all-

(preface- I am new to PI, as well as astrophotograpy in general)

I am having trouble every step of the way it seems.  Right now I am working with files from a Nikon d90, and tried to use the DSLR workflow.  I think that I figured out that one problem that I am having is that the bayer matrix for nikon GBRG rather than the default, so I tried to alter the workflow as such (basically I changed the BatchDebayer to use he GBRG), but that didn't seem to be enough (see the FITs that was integrated with calibration)

So, the first question is, for the DSLR workflow- what needs to be done differently for a nikon?

Since I think that my workflow with the calibration seems to be making things worse, I am abandoning that for now.

So, I changed the DSLR_RAW format back to disable the "Create RAW Bayer", and started re-processing from the NEF files-

I did the Alignment, then went on to integration- and at this point, I think that there is way to much garbage in my image.

I could go on with steps, but I think that I need a sanity check that the images that I am using are even worth trying things with.  Should there be this much blotchiness after integration?


(The tutorials aren't really helping with this because I am not even sure at which step I am going wrong, or how to alleviated it)

I put the fits files, as well as the raw NEFs, out on my dropbox.

The files are:
http://dl.dropbox.com/u/3523855/m82-fits/m82-integrated-from-calibrated.fit : the registered and integrated files calibrated with my (probably wrong) master bias & dark
http://dl.dropbox.com/u/3523855/m82-fits/m82-integrated-without-masters.fit : The registered and integrated files without calibration (which looks better but noisy?)
http://dl.dropbox.com/u/3523855/m82-fits/m82-RAW-Files.zip : Zipped Set of 8 raw NEF files

I have also included the process export(s) up through the integration step for both with and without calibration, just to see if anything jumps out at anyone.

Any Insight would be appreciated...

[pfile]

for starters, what is your focal ratio and subexposure length? how many subexposures are in the stack?

[JamieInCLT]

Right now, I am at f/10, and I have 8x30 second exposures.  Camera is a nikon d90, scope is a cpc-8" HD model.

Most of this is to learn the tools until my real camera & hyperstar come. 

I have a larger set of M42 that I can try with, but I haven't had a chance to grade those yet (still can't find that blink tool).  I will try that now...

[pfile]

okay. f/10 is quite slow and 30 seconds is very short. so you should not expect too much signal with only 4 minutes total integration time. that's probably the source of the quality issues you are seeing.

what do your back of camera histograms look like? for all CCDs, it's important for your signal to clear the read noise of your camera. a rule of thumb used for canon DSLRs is that all 3 channels of your histogram should be "well detached" from the left edge of the graph. some people say 20% across the way, some say 30% across the way... check out http://www.samirkharusi.net/ for a lot of DSLR-related astrophotography advice along the lines of optimum subexposure length and ISO.

[JamieInCLT]

okay. f/10 is quite slow and 30 seconds is very short. so you should not expect too much signal with only 4 minutes total integration time. that's probably the source of the quality issues you are seeing.

what do your back of camera histograms look like? for all CCDs, it's important for your signal to clear the read noise of your camera. a rule of thumb used for canon DSLRs is that all 3 channels of your histogram should be "well detached" from the left edge of the graph. some people say 20% across the way, some say 30% across the way... check out http://www.samirkharusi.net/ for a lot of DSLR-related astrophotography advice along the lines of optimum subexposure length and ISO.

Thanks for the feedback and the link- I added it to my "Reading List".  I will check out the histogram next time I am out.

I tried the same images in DSS, and ended up with a similar result- so you hit the mark- low signal to noise.

I ran through a batch of ~30x30s subs of m42 (also at f/10).  It looks much better.  I will post an image in the gallery, along with the workflow.  I *KNOW* there are kinks to work out but I am much happier with these results.

Thanks again!

[JamieInCLT]

It ALSO turns out that I messed up in acquisition, which might have something to do with why my calibrations failed.

I thought that I was at iso200, but it was really at 100.  I don't even know why I wasn't at least at 400 (the d90 gets a bit noisy over that)...

[pfile]

Thanks again!

yawelcome

It ALSO turns out that I messed up in acquisition, which might have something to do with why my calibrations failed.

I thought that I was at iso200, but it was really at 100.  I don't even know why I wasn't at least at 400 (the d90 gets a bit noisy over that)...

i can't seem to find any data for the d90 regarding it's so called 'unity gain'. this is the ISO where one electron causes the sensor output to increase by one. most of the time the ISO at unity gain is not one of the actual settings the camera has, so it's a theoretical number. shooting any higher than the unity gain represent a loss of dynamic range, so there's no reason to go above that point. shooting lower, while ok, can be a waste of sky time since your faint signal is going to be more likely to be swamped by read noise as the ISO goes down. you end up collecting many, many photons just to 'move the needle' on the sensor by one data number, and then for the weakest areas of signal, that DN is destroyed by the read noise just overwriting it. of course the dynamic range at lower ISO is higher, so there is a tradeoff here. but, i'm guessing that even ISO200 is probably much lower than the unity gain for your camera.

in areas of high light pollution, samir's advice is that lowering the ISO is okay, if at unity gain ISO your sensor saturates too quickly due to light pollution.