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

ASI071MC Pro default offset too high?

  • Please log in to reply
7 replies to this topic

#1 ZL4PLM

ZL4PLM

    Viking 1

  • *****
  • topic starter
  • Posts: 985
  • Joined: 23 Apr 2017
  • Loc: Darfield, New Zealand

Posted 16 February 2019 - 09:54 PM

I really struggled processing M42 on my first run with the ASI071MC Pro 

 

Seems the black level as waaaay too high - Pixinsight processes like ABE/DBE wouldn't process the data without creating wierd looking artifacts

 

I tweaked the black level (and of course lost data) but then could actually process correctly 

 

looking at the histo my black levels were well too far to the right 

 

I have the data there but tweaking the black level lost a lot of faint details

 

Whats the experience here?

 

I would say offset 20 or so could be a way better level 

 

this was shot at gain 90 / offset 65 

 

M42 ASI071MC Pro Gain 90 / Offset 65

 

Cheers

 

Simon 


  • elmiko likes this

#2 NorthField

NorthField

    Viking 1

  • *****
  • Posts: 811
  • Joined: 01 Jun 2017
  • Loc: SW Missouri

Posted 16 February 2019 - 11:18 PM

Iirc, the default/unity setting is 90 gain, 20 offset

All I ever used with it..
  • ZL4PLM likes this

#3 44maurer

44maurer

    Vostok 1

  • *****
  • Posts: 159
  • Joined: 12 Jun 2016
  • Loc: Los Angeles, CA

Posted 17 February 2019 - 12:05 AM

I follow the attached to determine my gain and offset.

 

http://www.astropixe..._and_offset.pdf



#4 bulrichl

bulrichl

    Explorer 1

  • -----
  • Posts: 90
  • Joined: 27 May 2018
  • Loc: La Palma (Canary Islands)

Posted 17 February 2019 - 05:43 AM

Hi Simon,

 

with my ASI071MC Pro at gain 90 the default offset value is 65, just as you wrote. This default setting is working perfectly for me.

 

The offset is meant to make sure that no intensity values are clipped to zero. This has to apply not only to light frames but also to dark frames and bias frames. Important: the check has to be done with bias frames since these are the frames that contain the lowest intensities.

 

An offset of 65 means that a constant value of 4 * 65 ADU = 260 ADU will be added to the original values. (The factor of 4 arises because the camera driver scales the intensity values of the 14-bit sensor to 16 bits.) So with the default setting (gain 90, offset 65) the mean of a bias frame should be near 260 ADUs. Did you check this?

 

Why do you regard this value to be too high? 260 / 65536 = 0.40 %  —  this is virtually nothing, you won't loose substantial dynamic range with this setting. In particular, you will not gain much by decreasing this offset value.

 

I guess you only took a look at your light frames and what you are seeing is a brightened sky background - this quite independent on the offset setting.

 

Bernd



#5 ZL4PLM

ZL4PLM

    Viking 1

  • *****
  • topic starter
  • Posts: 985
  • Joined: 23 Apr 2017
  • Loc: Darfield, New Zealand

Posted 17 February 2019 - 08:01 PM

as we live in a almost zero LP zone I'd be surprised if its a brightened sky background - our sky is prob some of the darkest in NZ :) 

 

the issues come in that the shoulder of the offset is well up the histogram with a large gap at 0 which when I stack this in pix the image is immensly green and bright - I can see colour in there but if I try using ABE then it just goes crazy. 

 

I tried SCNR first and remove the green and then run ABE but that gives me crazy swaths of pxielated colour and anything but an even background. To get anywhere near a chance to process it I had to adjust the black level and clip the bottom end and then I could run ABE etc and get sensible results which then gave me a fighting chance

 

I watched a video from Chris Gomez on the 071 and OSC pix processing and he simply stacked then - opened it and ran ABE then SCNR 

 

no chance!

 

to me the offset is too high the subs have too bright a background and needs to come down 

 

I saw a few people reported the same - gain 90 offset 20 seems to work

 

hence my asking what others were doing cos my processing isn't working as it seems!

 

:) 



#6 bulrichl

bulrichl

    Explorer 1

  • -----
  • Posts: 90
  • Joined: 27 May 2018
  • Loc: La Palma (Canary Islands)

Posted 18 February 2019 - 05:02 AM

Did you take some bias frames or very short dark frames (flat darks) and check the histogram and the statistics in PixInisght?

 

With my ASi071MC Pro (settings: gain 90, offset 65) the results looks like this,

 

a) for a flat dark (exposure time: 0.0055s):
"flat_dark.JPG"

 

b) for a light frame (exposure time: 300 s):
"light_frame.JPG"

 

Please note the different horizontal zoom in the input histograms. I am curious how that compares to your analogous frames.

 

Bernd

Attached Thumbnails

  • flat_dark.JPG
  • light_frame.JPG


#7 vdb

vdb

    Apollo

  • -----
  • Posts: 1376
  • Joined: 08 Dec 2009

Posted 18 February 2019 - 07:12 AM

Hmmm if you subtract the bias from the image, you should not be left with a bright background, I guess the issue is something else ...

 

/Yves



#8 Poochpa

Poochpa

    Surveyor 1

  • *****
  • Posts: 1755
  • Joined: 12 Feb 2005
  • Loc: Buffalo Burbs, NY

Posted 18 February 2019 - 11:37 AM

I'm using gain 50 with my 071P and default offset. Works well, but I'm using a fairly fast refractor, f/5.4.

Mike




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