Author Topic: The number of darks vs. bias and the use of darks  (Read 2547 times)

Offline Arnie

  • Newcomer
  • Posts: 5
The number of darks vs. bias and the use of darks
« on: 2015 November 29 02:51:24 »
First of all hello everyone, I am a newcomer both to the forums and have owned PI for abt 3 months. I am very pleased of the sofware and very eager to learn to use it.

I am a DSLR user (D7100) but am unaware of the details how the camera sensors work, ie how the picture is affected by different things in the macro level. So basicly I am very noob on that sense, so learning is almost 100 pct of following the tutorial examples and seeing what happens, then altering the parameters a bit and seeing again the difference. And although I can English pretty well the technical camera vocabularity is not that clear.

Before my first question this is a direct quote from PI main site on 'Master Calibration Frames: Acquisition and Processing' by Mr. Vicent Peris:

++++

IC is optimized to work with calibration libraries: don't worry about differing temperatures and exposure times between dark and light frames. IC will always rescale the dark noise to match every light frame.

It is also very important to have a bias library, because thermal noise must be bias-subtracted: only thermal noise must be rescaled in the master dark to match the thermal noise in the light frame.

After experimentation, I highly recommend taking at least 10 dark and bias frames. Never use less bias than dark frames, because this would introduce too much read noise in the bias-subtracted master dark: a 50-frame master dark will have almost all the read noise of a 10-frame master bias.

++++

Ok, I also just learned about making the super BIAS, and tested it also with one of my Master BIAS which was created from my abt 40 BIAS frames. The outcome was very smooth. But as the SUPER BIAS description tells me the program does something to the master BIAS so that it multiplies the Master by tens or even hundreds making it an equivalent of, let's say a couple of thousand individual BIAS frames. At least if there are the 200 BIAS frames to start with.

The thing that puzzles me is that as on the Mr. Peris tutorial he says that 'never use less BIAS than Darks'. I usually take 30 darks frames from each session. They are always the same length as my light frames and temperature is also quite close to the actual temp wich was outside as I start taking those outside when unriggin my gear.

Another thing I noticed now then is that PI is advanced software so it can scale the darks and that libraries are used. So basicly as I have gathered from all my sessions propably abt a few hundred darks during this year could I just use them now? But it would still be less than with the super BIAS.

I have not taken flats so far.

Here to recap my questions:
aa) The number of BIAS (using the super BIAS) vs the number of DARKs?
bb) How much does the noise differ on the same camera sensor with time?. (Ie could I just make a one Master Dark for each ISO setting I use out of for example thousand DARK frames of 10 sec exposure each taken indoors and use that always).
cc) Same as bb but with BIAS.

Thanks,

Arnie

Offline Arnie

  • Newcomer
  • Posts: 5
Re: The number of darks vs. bias and the use of darks
« Reply #1 on: 2015 November 30 05:52:12 »
All right, I thought I test this Bias/Dark library making and creating a Super Bias and Master Dark out of the frames.

I decided I do as per Light Vortex Astronomy tutorial. I took first 200 Bias frames at ISO 1600 and 1/8000 sec. I took them inside and put the interval to 10 seconds. Then I took 399 dark frames, each 10 sec exposure at ISO 1600 and intervall leaving 5 seconds in between in order for sensor to cool down a bit. Nor sure if it was enough as when looking at the histogram the closer to the 399 the pic number was the more reddish glowing it had. I assume that was due to the heat as the camera was in a totally dark place with no lense attached, lense mount capped obviously.

Then I created the Master BIAS with image integration, which I then run through the Super Bias script, and used the 'multiscale layers' value as '6' as per the tutorial.

After that I calibrated the individual dark frames with the super BIAS, and left the computer to start making the Master Dark with the Image Intergrationg script.

What I noticed was that the Super Bias did not look well in my opinion at all when stretched, at least compared to the one that I created out of another set of 40 individual BIAS frames. Stretching with the PI 24 bit stretch did not reveal anything, only the 'Ctrl + A' stretch did. I still did the dark frame calibration.

After doing the dark frame calibration with the Super Bias I opened the first dark frame as before and after the calibration. I zoomed in and to be honest I did not notice a single change on the pixel level. I will recheck when with the computer and post some pics.

I am just wondering if all this is more or less not worth doing? In terms of learning it definately is well used time but in terms of making a default Master Super Bias and Master Dark used with all ISO 1600 light frames in the future I am not so sure about...

Arnie

Offline Arnie

  • Newcomer
  • Posts: 5
Re: The number of darks vs. bias and the use of darks
« Reply #2 on: 2015 November 30 09:45:37 »
All right, computer had done the Master Dark out of the calibrated frames. I also checked closer all the phases and noticed that indeed there are differences also between the individual dark and calibrated dark. I also noticed that the Super Bias was actually much nicer looking when stretched with the STF.

Please take a look at the outcome. All I did with the pics ws the STF, then I saved them JPEG format, took them through LR to get them smaller in size.

I do not know now then if this Master Dark less Super Bias is well formed or not. The initial dilemma I still have is that can I now use this Master Dark and Super Bias as my 'ISO 1600' default Masters? And if so, are they more or less permanent or should I remake one after a certain period of time?

Here they are as DropBox files:
-Master Bias (200 individuals): https://dl.dropboxusercontent.com/u/38951141/PI%20FORUM/MASTER%20BIAS%20200%20ISO%201600%20FRAMES-1.JPG
-Super Bias: https://dl.dropboxusercontent.com/u/38951141/PI%20FORUM/SUPER%20BIAS%20200%20ISO%201600-1.JPG
-10 sec Individual Dark: https://dl.dropboxusercontent.com/u/38951141/PI%20FORUM/10%20sec%20individual%20Dark%20ISO%201600-1.JPG
-10 sec calibrated individual Dark: https://dl.dropboxusercontent.com/u/38951141/PI%20FORUM/10%20sec%20individual%20calibrated%20Dark%20ISO%201600-1.JPG
-Master Dark ISO 1600: https://dl.dropboxusercontent.com/u/38951141/PI%20FORUM/MASTER%20DARK%20LESS%20SUPER%20BIAS%20ISO%201600-1.JPG

Please I would appreciate a comment.

-Arnie