Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Master dark is too dark!

  • Please log in to reply
15 replies to this topic

#1 descott12

descott12

    Viking 1

  • -----
  • topic starter
  • Posts: 639
  • Joined: 28 Sep 2018
  • Loc: Charlotte, NC

Posted 21 April 2019 - 11:00 AM

Hello all,

I do EAA but I thought the masters in this forum may have a little more experience with this. I have an ASI 294 MC Pro and I use SharpCap. I have had good results with flats so I attempted to make some darks.

 

Raw 16, 4144x2822, Gain 200, Exposure 10 and 15 seconds, temp set to 5C

 

I averaged 10 frames and the resulting dark is just uniform almost black. No hot pixels, amp glow. Nothing.

 

I guess I should be happy that my camera is so perfect! but I can't help but wonder if I didn't do something correctly. SC makes it pretty idiot-proof so I am not sure where I could have messed up.

 

If I understand darks, there is not point in subtracting a dark that has basically zero noise.

Thanks in advance.


Edited by descott12, 21 April 2019 - 11:44 AM.


#2 RedLionNJ

RedLionNJ

    Gemini

  • *****
  • Posts: 3305
  • Joined: 29 Dec 2009
  • Loc: Red Lion, NJ, USA

Posted 21 April 2019 - 11:43 AM

Hi Dave,

 

Is it still truly uniform (at zero) all the way across the field even if you execute the equivalent of a STF (in PixInsight)?

 

If that's truly the case, then the dark adds no value.  I find this incredibly difficult to accept, though.

 

It's only when we fully-calibrate and then stretch the bejesus out of data that the value of a good set of averaged darks (or flats) comes into play.



#3 Gipht

Gipht

    Surveyor 1

  • *****
  • Posts: 1617
  • Joined: 12 Nov 2016
  • Loc: Prescott Valley, AZ.

Posted 21 April 2019 - 12:18 PM

Is there a problem with the finished photograph?  Sensor glow showing for example?  



#4 descott12

descott12

    Viking 1

  • -----
  • topic starter
  • Posts: 639
  • Joined: 28 Sep 2018
  • Loc: Charlotte, NC

Posted 21 April 2019 - 12:43 PM

Hi Dave,

 

Is it still truly uniform (at zero) all the way across the field even if you execute the equivalent of a STF (in PixInsight)?

 

If that's truly the case, then the dark adds no value.  I find this incredibly difficult to accept, though.

 

It's only when we fully-calibrate and then stretch the bejesus out of data that the value of a good set of averaged darks (or flats) comes into play.

 

Thanks for the reply.

It is not zero, just a very dark grey. The histogram shows a spike all the way to the left so there is some data just very low level and it does seem to be very uniform. I really expected to see some sort of noise or a hot pixel or two. Maybe a 15 second exposure is not long enough? I know you AP people typically do much longer exposures.

 

As I am not an AP guy, I am not sure what an STF is and I don't use PI. But I do know how to get around in Gimp so maybe there is something I could look for? I didn't bother uploading the dark as I thought that shrinking it down to a small enough size would destroy the data



#5 descott12

descott12

    Viking 1

  • -----
  • topic starter
  • Posts: 639
  • Joined: 28 Sep 2018
  • Loc: Charlotte, NC

Posted 21 April 2019 - 12:44 PM

Is there a problem with the finished photograph?  Sensor glow showing for example?  

Not at all but many in the EAA forum have reported seeing benefits with a dark frame so I figured I would give it a try. Flats definitely work well. But maybe I am trying to fix something that "ain't broke" as the saying goes.



#6 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 6897
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 21 April 2019 - 01:32 PM

You have to stretch the dark to see it properly. Otherwise it just looks black. Stretch the histogram aggressively, and details will emerge.


  • bobzeq25 and DSO_Viewer like this

#7 spokeshave

spokeshave

    Surveyor 1

  • *****
  • Posts: 1876
  • Joined: 08 Apr 2015

Posted 21 April 2019 - 03:33 PM

It's all there, you just need to stretch the darks to see the noise and hot pixels. Remember, darks are photographs of utter blackness. One would expect them to appear black in the linear state. 

 

Also, hot pixels can be difficult to see unless you zoom in quite a bit. Open up a dark frame in Gimp and use the "Levels" slider to brighten the image. The noise will start to emerge and if you zoom in a bit, you'll see the hot pixels.

 

Tim



#8 descott12

descott12

    Viking 1

  • -----
  • topic starter
  • Posts: 639
  • Joined: 28 Sep 2018
  • Loc: Charlotte, NC

Posted 21 April 2019 - 04:03 PM

Hey guys,

Here is an attached copy that is stretched about as much as is possible without turning it white. You can see there are some spots that sort of look like something you would see on a flat but nothing that looks like a hot pixel. What do you think?  Or am I just expecting the wrong appearance? 

Attached Thumbnails

  • Dark_Stretched.png

Edited by descott12, 21 April 2019 - 05:29 PM.


#9 Alex McConahay

Alex McConahay

    Cosmos

  • *****
  • Posts: 7672
  • Joined: 11 Aug 2008
  • Loc: Moreno Valley, CA

Posted 21 April 2019 - 05:55 PM

Flats are not really about hot pixels so much as uneven illumination. 

 

ALex



#10 descott12

descott12

    Viking 1

  • -----
  • topic starter
  • Posts: 639
  • Joined: 28 Sep 2018
  • Loc: Charlotte, NC

Posted 21 April 2019 - 06:10 PM

Flats are not really about hot pixels so much as uneven illumination. 

 

ALex

Yes, exactly. This is a dark and it looks like a flat!



#11 spokeshave

spokeshave

    Surveyor 1

  • *****
  • Posts: 1876
  • Joined: 08 Apr 2015

Posted 21 April 2019 - 06:34 PM

First thing - the spots are dust shadows and you should not be able to see them in darks. It means that you have a light leak. It takes light to cast a shadow. How are you taking you darks?

 

Second - it will be difficult if not impossible to detect hot pixels looking at the whole image, particularly after it has been resized to fit the CN posting limits. As I mentioned above, you need to zoom in to see them. I downloaded your image and zoomed in in PS and sure enough, there are hot pixels all over the place.

 

Tim


  • DSO_Viewer likes this

#12 descott12

descott12

    Viking 1

  • -----
  • topic starter
  • Posts: 639
  • Joined: 28 Sep 2018
  • Loc: Charlotte, NC

Posted 21 April 2019 - 07:18 PM

Hi Tim,

Thanks for checking. I took the darks with with the front cap screwed on and the camera was wrapped inside a thick towel to block ambient light from the side vents. I am not sure how I could make it darker - I guess I will try again with the room light off but I really thought I had it covered well.

 

I zoomed into that image to the max with Gimp and another viewer and didn't see anything. Maybe I don't know what a hot pixel should look like - I assumed it would be a small bright spot but I am not seeing anything like that.  Sorry for wasting your time...



#13 spokeshave

spokeshave

    Surveyor 1

  • *****
  • Posts: 1876
  • Joined: 08 Apr 2015

Posted 21 April 2019 - 07:51 PM

Not wasting my time, Dave. I'm glad to help. I circled some of the hot pixels. You can see that I needed to zoom in to the point where individual pixels are visible:

 

hotpixels.PNG

 

Try taking the darks in a dark room. Light is getting in from somewhere and it doesn't take much.

 

Tim


Edited by spokeshave, 21 April 2019 - 07:51 PM.

  • DSO_Viewer likes this

#14 descott12

descott12

    Viking 1

  • -----
  • topic starter
  • Posts: 639
  • Joined: 28 Sep 2018
  • Loc: Charlotte, NC

Posted 21 April 2019 - 08:43 PM

Ah yes, I see them. I guess I was expecting something much brighter. Thanks again.



#15 OldManSky

OldManSky

    Viking 1

  • -----
  • Posts: 865
  • Joined: 03 Jan 2019
  • Loc: Valley Center, CA USA

Posted 21 April 2019 - 08:57 PM

Dave, I notice you said you were doing EAA, but were taking 15 second darks...are you doing 15 second exposures for your lights.

You know the dark exposure time needs to match your lights, right.

Just checking smile.gif



#16 descott12

descott12

    Viking 1

  • -----
  • topic starter
  • Posts: 639
  • Joined: 28 Sep 2018
  • Loc: Charlotte, NC

Posted 21 April 2019 - 09:19 PM

Dave, I notice you said you were doing EAA, but were taking 15 second darks...are you doing 15 second exposures for your lights.

You know the dark exposure time needs to match your lights, right.

Just checking smile.gif

Yes, I usually use a 10 or 15 second exposure so I made darks with both of those exposures.




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics