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

gain and offset settings for ZWO cameras

  • Please log in to reply
29 replies to this topic

#1 Organic Astrochemist

Organic Astrochemist

    Viking 1

  • *****
  • topic starter
  • Posts: 863
  • Joined: 10 Jan 2015

Posted 30 June 2016 - 12:09 PM

My latest learning/amusement with my ASI178MM-Cool camera is the result of experimenting with the gain and offset settings. Does anyone have any advice or experience in optimizing these settings for ZWO cameras?

In my case the camera has 15K full well depth and 2^14 = 16 k bit depth, so initially I left the gain at 0 and adjusted the exposure time. But two nights ago I increased the gain to 375 and shot 1000 subs at 3 seconds. I think I went deeper than ever before (I was also clipping the brightest stars). So perhaps previously I wasn't filling my well up very efficiently.

Messier 4

How can I know what the pixel values are? In a given image, how many pixels should be at max (clipped)? Is the histogram a good enough guide (peak at 1/4 to 1/3 from the left) and not touching/barely touching the right?

 

I also increased the offset. Even though I shoot in a white zone, due to the small pixel size and short exposures and quantization error some of my pixels might still read zero. Does it make sense to increase the offset? How can I optimize this? Again, does the histogram serve as a guide (not touching the left)?

Thanks for your invaluable advice and experience.



#2 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23290
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 30 June 2016 - 03:02 PM

To properly determine the necessary offset, it helps to have accurate statistical evaluation of every pixel in the image. A histogram can help, but it is really not the best tool. Personally I use the Statistics tool in PixInsight, which will give me a proper statistical evaluation of all the pixels of an image. I also like to use PixelMath to render clipped to black pixels as white, and all non-clipped pixels as black, and I can then run other evaluations on the resulting image to determine clipped pixel counts, etc. 

 

A proper offset will give you zero clipped to black pixels in every test frame. For a statistically relevant evaluation, you will need to test a set of frames once you think you have found the setting that will avoid any clipped pixels at all. The idea here is that the offset is just large enough to avoid any clipping of any pixels due to noise, but not larger than that, which would needlessly cost you dynamic range.



#3 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • Posts: 2989
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 30 June 2016 - 10:11 PM

Hi John.

Can you post your suggested findings for gain and offset for the 1600. Maybe others can post gain and offset for other ZWO cameras. This way this thread can be valueable to future ZWO camera buyers

As I recall for the ASI 1600 with unity gain 139 an offset of 21 is used. Unity gain for the ASI 174 is I believe 179 with offset of 46

Al

#4 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23290
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 30 June 2016 - 10:34 PM

With the ASI1600, the ASCOM driver has built-in presets for low, unity and high gain that automatically sets the offset as well. Low uses an offset of 10, unity an offset of 21 and high an offset of 50. I settled on a 12-14 offset for Gain 75, which I also use. I have not done testing at other gains, so I can't offer much there. 



#5 Easton7172001

Easton7172001

    Lift Off

  • *****
  • Posts: 21
  • Joined: 13 Jan 2014

Posted 15 February 2017 - 03:41 PM

I just gor my Zwo asi1600mmc and i wonder if anybody has done live view on this camera? If so what setting are you using cause i am having a hard time aligning my scope.
Thank

#6 A. Viegas

A. Viegas

    Mercury-Atlas

  • *****
  • Posts: 2989
  • Joined: 05 Mar 2012
  • Loc: New York City/ CT

Posted 15 February 2017 - 05:54 PM

There are many people using the color version of this camera for live view. I suggest you log into www.nightskiesnetwork.com it's free and most evenings there will be people broadcasting and if someone is using the 1600mc you can ask questions and see live view changes with different settings

I think when I have used my 1600mm with sharpcap for live viewing I used 300 gain.

Al
  • ABQtom likes this

#7 Thirteen

Thirteen

    Gemini

  • *****
  • Posts: 3156
  • Joined: 12 Jul 2013
  • Loc: Milford, Michigan

Posted 15 February 2017 - 10:01 PM

Some time ago I plotted out the default gain and offset options that ZWO offered in the ASCOM driver and then just drew a curve through them.    I honestly think you would be fine just figuring an offset to use from the chart by finding your gain and reading across to the offset to use.   You could definitely dive into the statistics if you want, but I would be surprised if it ended up significantly different than this.   Also, if you are concerned you could cheat the offset selection a little on the high side of the curve and not have disastrous effects.

 

 

Gain and Offset.jpg


Edited by Thirteen, 15 February 2017 - 10:02 PM.

  • calan, Jon Rista, ChrisWhite and 2 others like this

#8 vagrom

vagrom

    Vostok 1

  • -----
  • Posts: 190
  • Joined: 05 Sep 2016
  • Loc: Kansas City

Posted 16 February 2017 - 09:04 AM

 You could definitely dive into the statistics if you want, but I would be surprised if it ended up significantly different than this.  

 

Thanks for the graph. If one would want to dive into the statistics, where would one learn more?


  • Scott1244 likes this

#9 Seanem44

Seanem44

    Vanguard

  • *****
  • Posts: 2007
  • Joined: 22 Sep 2011
  • Loc: Stafford, VA

Posted 04 April 2017 - 11:09 AM

I'm bumping this back up into awareness since I just purchases a 1600MM-C.  Being a former DSLR user, this is still so foreign to me, so I am trying to give myself a leg up.

 

Can anyone else add to this?


  • Scott1244 likes this

#10 f430

f430

    Messenger

  • *****
  • Posts: 420
  • Joined: 29 Sep 2015
  • Loc: La Mesa, CA.

Posted 04 April 2017 - 11:56 AM

I just gor my Zwo asi1600mmc and i wonder if anybody has done live view on this camera? If so what setting are you using cause i am having a hard time aligning my scope.
Thank

I realize this is an old question, but I use my 1600mmc camera to align my scope. I use Nebulosity, set the camera settings at about 250 gain/ about 25 offset and 1 or 2 second exposure. (just enough gain to be able to see stars clearly with a quick exposure for the alignment) and use the Frame and Focus setting on Nebulosity. Works great! With a bullseye to center the star in.

John


Edited by f430, 04 April 2017 - 11:57 AM.


#11 Don Walters

Don Walters

    Mariner 2

  • -----
  • Posts: 207
  • Joined: 06 May 2015

Posted 04 April 2017 - 05:13 PM

From suggestions I've gotten on these forums, I am using the driver presets for my 178MM-Cool.   I picked "low read noise" option in the driver presets and use exposure times of about 2 minutes for narrowband, and about 30 seconds for RGB.  I'm not as thorough as Jon and some of the others here, so I don't know if I'm really doing this properly (considering my choices for exposure times) but my images are coming out pretty nice, so I'm happy for now until I decide to start splitting hairs.



#12 calan

calan

    Viking 1

  • *****
  • Posts: 805
  • Joined: 16 Jun 2007
  • Loc: Oklahoma City, OK

Posted 01 May 2017 - 06:36 PM

Some time ago I plotted out the default gain and offset options that ZWO offered in the ASCOM driver and then just drew a curve through them.    I honestly think you would be fine just figuring an offset to use from the chart by finding your gain and reading across to the offset to use.   You could definitely dive into the statistics if you want, but I would be surprised if it ended up significantly different than this.   Also, if you are concerned you could cheat the offset selection a little on the high side of the curve and not have disastrous effects.

 

 

attachicon.gifGain and Offset.jpg

Jason,

 

Do you happen to have a chart like this for the ASI224MC?



#13 Thirteen

Thirteen

    Gemini

  • *****
  • Posts: 3156
  • Joined: 12 Jul 2013
  • Loc: Milford, Michigan

Posted 01 May 2017 - 09:04 PM

I don't. I made it by just plotting the preset values as they appear in the ASCOM driver.
  • calan and keithlt like this

#14 keithlt

keithlt

    Surveyor 1

  • *****
  • Posts: 1579
  • Joined: 21 Dec 2013
  • Loc: Arizona©

Posted 01 May 2017 - 09:29 PM

I don't. I made it by just plotting the preset values as they appear in the ASCOM driver.

I would like to see this chart plotted for the qhy163mm. Oh that would be cool.



#15 calan

calan

    Viking 1

  • *****
  • Posts: 805
  • Joined: 16 Jun 2007
  • Loc: Oklahoma City, OK

Posted 04 May 2017 - 01:33 AM

I plotted a chart for the ASI224MC, based on ZWO's presets (in pink), and a linear interpolation to get the other offset values. I'll post it here for future reference.

The last column is the ideal target ADU value, based on Jon's 20xRN formula. There is tons of good info on that in this thread. The rest of the data came from the charts posted on ZWO's website.

ASI224MC_numbers.jpg

 


Edited by calan, 04 May 2017 - 01:40 AM.

  • StevenBellavia, cajusor and Starman08 like this

#16 StevenBellavia

StevenBellavia

    Ranger 4

  • *****
  • Posts: 376
  • Joined: 11 Mar 2014
  • Loc: New York

Posted 13 December 2017 - 03:37 PM

I plotted a chart for the ASI224MC, based on ZWO's presets (in pink), and a linear interpolation to get the other offset values. I'll post it here for future reference.

The last column is the ideal target ADU value, based on Jon's 20xRN formula. There is tons of good info on that in this thread. The rest of the data came from the charts posted on ZWO's website.

attachicon.gifASI224MC_numbers.jpg

 

Hi Calan,

I took your lovely chart and tried to re-calculate the ZWO Gain numbers (which I think I succeeded) and also added a column for those of us still stuck in the DSLR world (anyone remember ISO and DSLR cameras here?)

I also changed the e/ADU in the higher gains, so it would match the calculation using decibels, which is what ZWO says they do, per this thread on their forum:

 

http://zwoug.org/vie...php?f=21&t=6154

 

I do not claim this to be 100% accurate.  Just a guide.

Feel free to comment (but not insult)

 

Steve

Attached Thumbnails

  • ZWO_ASI_224M_Gain-decibels-ISO.JPG

Edited by StevenBellavia, 13 December 2017 - 03:38 PM.

  • mjcastles likes this

#17 StevenBellavia

StevenBellavia

    Ranger 4

  • *****
  • Posts: 376
  • Joined: 11 Mar 2014
  • Loc: New York

Posted 15 December 2017 - 02:08 PM

 

I plotted a chart for the ASI224MC, based on ZWO's presets (in pink), and a linear interpolation to get the other offset values. I'll post it here for future reference.

The last column is the ideal target ADU value, based on Jon's 20xRN formula. There is tons of good info on that in this thread. The rest of the data came from the charts posted on ZWO's website.

attachicon.gifASI224MC_numbers.jpg

 

Hi Calan,

I took your lovely chart and tried to re-calculate the ZWO Gain numbers (which I think I succeeded) and also added a column for those of us still stuck in the DSLR world (anyone remember ISO and DSLR cameras here?)

I also changed the e/ADU in the higher gains, so it would match the calculation using decibels, which is what ZWO says they do, per this thread on their forum:

 

http://zwoug.org/vie...php?f=21&t=6154

 

I do not claim this to be 100% accurate.  Just a guide.

Feel free to comment (but not insult)

 

Steve

 

I meant log(GADU at 0/GADU)

Attached Thumbnails

  • ZWO_ASI_224M_Gain-decibels-ISO.JPG


#18 Organic Astrochemist

Organic Astrochemist

    Viking 1

  • *****
  • topic starter
  • Posts: 863
  • Joined: 10 Jan 2015

Posted 17 December 2017 - 11:54 PM

It's been a while since I checked in on this thread, but I followed the lead of others and plotted the ASCOM driver gain vs offset for the ASI178MM. I think this might be important for people (like me) with this camera. I think I've been setting the offset too low.

Thanks for all the contributions.

offset graph.jpg



#19 StevenBellavia

StevenBellavia

    Ranger 4

  • *****
  • Posts: 376
  • Joined: 11 Mar 2014
  • Loc: New York

Posted 09 January 2018 - 12:17 AM

I just spent many hours with my new QHY5-III-178 color camera, to get the "correct" offsets for various gain settings.

 

The ASCOM driver only has one choice, which is "DSO", and I believe it is wrong (Gain 15, offset 307).

 

This is what I come up with, using the criteria that the minimum value of all pixels is greater than 100 and less than 1,000, per Craig Stark:

 

http://www.stark-lab...inAndOffset.php

 

Quick summary:

 

Gain   Offset

  0         20

10         40

15         65

20        110

30        320

51      1023

 

Note, the QHY users manual shows the gain going up to 65, but 51 is the limit with the ASCOM driver, and 1023 the offset limit

 

These were all done at 25C.  Not sure how much it would change at other temperatures.

 

Attached Thumbnails

  • QHY5-III-178_Offset_vs_Gain.PNG


#20 StevenBellavia

StevenBellavia

    Ranger 4

  • *****
  • Posts: 376
  • Joined: 11 Mar 2014
  • Loc: New York

Posted 09 January 2018 - 09:23 AM

New and improved:

 

I don't think Gain 51 is of any use.

Even at Gain 40, the maximum offset still clips.

Latest settings:

 

Gain   Offset

  0         20

10         40

15         65

20        110

30        320

40      1023

 

Now my next issue:  I can't get the sensor temperature out.  Is it not available with this series or just bad EXIF data?  I've submitted a ticket to QHY, so we will see.  APT says 25C for all my images, but I suspect they were actually much warmer.  And this does affect the offsets.  Colder temps would require a higher offset, etc.

Attached Thumbnails

  • QHY5-III-178_Offset_vs_Gain_A.PNG


#21 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23290
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 09 January 2018 - 11:30 AM

 And this does affect the offsets.  Colder temps would require a higher offset, etc.

 

Why would colder temps require a higher offset? The offset is intended to prevent pixels from clipping to black due to noise. At higher temps, you have more noise. At colder temps, you have less noise. It therefor follows that you would not need a larger offset with a colder sensor, since the total noise would be lower. 



#22 AstroPics

AstroPics

    Ranger 4

  • *****
  • Posts: 309
  • Joined: 11 Jan 2017
  • Loc: Atlanta, GA

Posted 09 January 2018 - 11:42 AM

Some time ago I plotted out the default gain and offset options that ZWO offered in the ASCOM driver and then just drew a curve through them.    I honestly think you would be fine just figuring an offset to use from the chart by finding your gain and reading across to the offset to use.   You could definitely dive into the statistics if you want, but I would be surprised if it ended up significantly different than this.   Also, if you are concerned you could cheat the offset selection a little on the high side of the curve and not have disastrous effects.

 

 

attachicon.gifGain and Offset.jpg

This is interesting and somewhat related to another thread (related to SGP only being able to set gain and not offset).

 

https://www.cloudyni...s/#entry8326017

 

So, does this mean if you use the ZWO ASCOM driver and only set the gain, you will get a default offset setting as well from the chart above?



#23 StevenBellavia

StevenBellavia

    Ranger 4

  • *****
  • Posts: 376
  • Joined: 11 Mar 2014
  • Loc: New York

Posted 13 February 2018 - 02:00 PM

 

 And this does affect the offsets.  Colder temps would require a higher offset, etc.

 

Why would colder temps require a higher offset? The offset is intended to prevent pixels from clipping to black due to noise. At higher temps, you have more noise. At colder temps, you have less noise. It therefor follows that you would not need a larger offset with a colder sensor, since the total noise would be lower. 

 

Thanks Jon.  I had that backwards (or just plain wrong)

While I "have" you - would you know the equivalent gain settings from your ZWO ASI183 for a QHY 183?  I know you said you liked gain 53 on the ZWO for that camera and I am trying to find a good starting point for the QHY 183.

 

Thanks again,

Steve


Edited by StevenBellavia, 13 February 2018 - 02:00 PM.


#24 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23290
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 13 February 2018 - 03:41 PM

I don't know exactly the gain settings for the QHY. I suspect it is in the vicinity of 40-50, but I'd need to test to know the exact gain. Being exact isn't all that important...I'm just a bit of a nut when it comes to knowing my read noise and conversion ratio and whatnot, which is why I pick exact settings like 53, rather than just "about 50".
  • StevenBellavia likes this

#25 StevenBellavia

StevenBellavia

    Ranger 4

  • *****
  • Posts: 376
  • Joined: 11 Mar 2014
  • Loc: New York

Posted 13 February 2018 - 04:39 PM

 

 

 And this does affect the offsets.  Colder temps would require a higher offset, etc.

 

Why would colder temps require a higher offset? The offset is intended to prevent pixels from clipping to black due to noise. At higher temps, you have more noise. At colder temps, you have less noise. It therefor follows that you would not need a larger offset with a colder sensor, since the total noise would be lower. 

 

Thanks Jon.  I had that backwards (or just plain wrong)

While I "have" you - would you know the equivalent gain settings from your ZWO ASI183 for a QHY 183?  I know you said you liked gain 53 on the ZWO for that camera and I am trying to find a good starting point for the QHY 183.

 

Thanks again,

Steve

 

I think I got it. 

 

I got this curve from the QHY manual, and it seems, like ZWO, their gain numbers do correspond to decibels of e/ADU, but directly (not times ten).

 

So I think I answered my own question:

 

The QHY183 gain of 5 (2e/ADU) is similar to the ZWO183 gain of 53.

Attached Thumbnails

  • QHY_System_gain.JPG
  • ZWO_183_and_QHY_183_Gain.JPG

Edited by StevenBellavia, 14 February 2018 - 12:57 PM.



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