Author Topic: Superbias vs. SuperBias-pxm.dll [PI 1.8]  (Read 4405 times)

Offline ArminPressler

  • Newcomer
  • Posts: 4
    • Armins Portfolio
Superbias vs. SuperBias-pxm.dll [PI 1.8]
« on: 2015 April 13 09:16:53 »
Hello!
I own PI since February and I am infected  >:D

After a long and hard learning periode my personal workflow in PI is almost done. (Yes, at the end I will say '100% in PI').

But I am still reading and learning, so I found an article describing a 'SuperBias' process.
(http://trappedphotons.com/blog/?p=693) and (http://pixinsight.com/forum/index.php?topic=7312)

So I was downloading the *.dll from http://www.astrophoto.cl/Research.html and did not realize, that there is in PI 1.8 already included a 'Superbias' process.  ::)

I'm not sure, if my Moravian 8300 mono (KAF-8300) needs the Superbias process, if I over stretch the Masterbias I can see very small and tiny vertical lines...


The two Superbias moduls have two different process windows and the result is also different:


Which one of the processes do you recommend, the internal or the external?

Thanks,
Armin




Offline Carlos Milovic

  • PTeam Member
  • PixInsight Jedi Master
  • ******
  • Posts: 2172
  • Join the dark side... we have cookies
    • http://www.astrophoto.cl
Re: Superbias vs. SuperBias-pxm.dll [PI 1.8]
« Reply #1 on: 2015 April 13 10:47:05 »
The "internal" one is an upgrade that was done by Juan over my "external" code. It should run a bit faster, and is more flexible. It works on views, while my code worked on files. Other than that, it should be almost the same.
(Having said that, I'll have to check that "Exclude large-scale structure" option... it should be the other way around, discarding small scale structures).
Regards,

Carlos Milovic F.
--------------------------------
PixInsight Project Developer
http://www.pixinsight.com

Offline ArminPressler

  • Newcomer
  • Posts: 4
    • Armins Portfolio
Re: Superbias vs. SuperBias-pxm.dll [PI 1.8]
« Reply #2 on: 2015 April 13 11:09:56 »
Hi Carlos!

Thanks for your quick reply and your explanations!

I presume, that in my case I don't need the SuperBias process?
Or should I fine tune the parmaters to get a smoother SuperBias?

Offline Carlos Milovic

  • PTeam Member
  • PixInsight Jedi Master
  • ******
  • Posts: 2172
  • Join the dark side... we have cookies
    • http://www.astrophoto.cl
Re: Superbias vs. SuperBias-pxm.dll [PI 1.8]
« Reply #3 on: 2015 April 13 12:05:25 »
There are some vertical patterns that may benefit from SB. Anyway, you should inspect the bias at 1:1 and using the 24bit STF. A normal STF will show you artifacts due to polarization.
I think that you'll get smoother results with the old module.
Regards,

Carlos Milovic F.
--------------------------------
PixInsight Project Developer
http://www.pixinsight.com

Offline ArminPressler

  • Newcomer
  • Posts: 4
    • Armins Portfolio
Re: Superbias vs. SuperBias-pxm.dll [PI 1.8]
« Reply #4 on: 2015 April 13 12:43:33 »
Carlos, Thanks for your help!

To be honest, I am not really know what I am doing in this case  :-[

It is the 1st time that I use 24Bit LUT - I had to read in the tutorial where to find the button (https://pixinsight.com/tutorials/24-bit-stf/)

So I push the 24Bit LUT button and applied a STF auto stretch:


the question is, is the superbias picture 'better' than the masterbias ?

I think, I should always use the 24-Bit LUT in my workflow?

Thanks,
Armin


Offline TobiasLindemann

  • PixInsight Addict
  • ***
  • Posts: 126
    • Trackingstation
Re: Superbias vs. SuperBias-pxm.dll [PI 1.8]
« Reply #5 on: 2015 April 13 13:21:40 »
Hi Armin,

the 24bit mode is only for correct visualisation and has no effect to the image.
As you see, the superbias is much more smooth and has no noise inside. The lines you see, are the real signal of the bias, that is caused by the readout.
I always use superbias and it works great for me.

Greetings
Tobias

Offline ArminPressler

  • Newcomer
  • Posts: 4
    • Armins Portfolio
Re: Superbias vs. SuperBias-pxm.dll [PI 1.8]
« Reply #6 on: 2015 April 13 23:22:47 »
Hi Tobias!

Thank you for your reply!
In future I also will use the SuperBias instead if the MasterBias :-)

I knew, that the 24Bit mode do not effect the picture itself, but I think in some cases it would be good to have a correct visualisation.
Maybee if the picture itself is 64Bit or so? If I stack more than -I don't knew- 50 to 100 lights, what do you think?


Offline Carlos Milovic

  • PTeam Member
  • PixInsight Jedi Master
  • ******
  • Posts: 2172
  • Join the dark side... we have cookies
    • http://www.astrophoto.cl
Re: Superbias vs. SuperBias-pxm.dll [PI 1.8]
« Reply #7 on: 2015 April 14 07:07:16 »
It depends on the dynamic range and the noise level. I often see that I need a 24bits STF after noise reduction in linear mode, or to inspect a background created by DBE or ABE.


Truth to be told, the gain using SB is not going to be dramatic. Sure, it incorporates less noise to the lights, but the amplitude is so low, that only marginal differences are going to show in the final master light. Since it is so easy and fast to do, I'd rather use it that leave it. I always used SB with my Canon 300D. Now my T2i doesn't show any vertical pattern, so I don't use it anymore with my data.
Regards,

Carlos Milovic F.
--------------------------------
PixInsight Project Developer
http://www.pixinsight.com