Binning ASI120MC

Anything that doesn't fit into any of the other forums
Post Reply
User avatar
oopfan
Posts: 105
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Binning ASI120MC

Post by oopfan » Mon Jul 24, 2017 9:34 pm

Thank you for that great tutorial on Binning!

The user manual for my ASI120MC says this under the heading named "Binning":
The ASI120 camera supports software bin2 mode. You should use 10bit ADC. We recommend customer to use software binning if you don't care speed.
Would you be kind enough to translate that into everyday English?

Thank you.

User avatar
oopfan
Posts: 105
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Binning ASI120MC

Post by oopfan » Tue Jul 25, 2017 3:00 pm

Let me be more specific. When ZWO talks about "software" do they mean the software in SharpCap, or the Windows driver for the camera, or ZWO's firmware, or software incorporated in the sensor? I mean, software has its hands in the magic from beginning to end. From my perspective, SharpCap gives me the option of "2x2" binning. So who is responsible for performing that magic? The ZWO company has a lot of smart people on staff but their docs need improvement. The way it is written they give me the impression that binning can happen two places: on their side or on our side (i.e. SharpCap). Is it just me, or are they telling me not to use ZWO's binning? Why would that be?

User avatar
admin
Site Admin
Posts: 820
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Binning ASI120MC

Post by admin » Tue Jul 25, 2017 3:41 pm

As I mentioned in my long, rambling, post on binning, true Hardware binning (where the charge from different pixels is summed before A->D conversion) only happens on CCD cameras. All other binning is software binning of one sort or another.

In the ZWO case (and for any other CMOS camera that has a so-called 'hardware binning' option) there are two places the summing or averaging of pixel values can be done

1) On the camera itself
2) In the Camera SDK or driver on the PC

Option 2 is very boring - everything proceeds as for capturing at full resolution until the frame is in the driver/SDK and then some code running on the PC turns it into a smaller frame by averaging or adding pixels.

Option 1 typically has some tradeoffs... Since the binning is happening on the device itself, CPU usage on the PC may be lower, but this is usually a trivial effect. Also since less data needs to be transferred from the camera to the PC you *may* get higher frame rates if the USB bandwidth was limiting your frame rates. Finally, depending on the sensor model, enabling this often has the side effect of using the ADC in a different mode. I think it's this side effect that ZWO are warning against. Off the top of my head I can't quite recall what the change to the ADC is, but I think it may increase the noise level maybe?

Hope that helps,

Robin

User avatar
oopfan
Posts: 105
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Binning ASI120MC

Post by oopfan » Fri Jul 28, 2017 1:35 am

Robin,

You might find this interesting. The last time I had some good weather was July 15th (12 days ago)! I had one last shot at capturing the Dumbbell. My motto is "do more with less" so I refuse to do guided astrophotography what with the cost of a guide scope and camera. Call me cheap. I was testing out my new modification to my custom RA stepper motor drive that quantum-leaped me from 30 sec subframes (max) to 120 sec subframes. I am hoping to get that to 300 sec with my next version of software for Periodic Error Correction. Anyhow I do my imaging with a Celestron TravelScope 70, an $80 value, and ASI120MC. I knew from experience that my pixel scale was 1.933 arcsec / pixel. I was only able to capture 7 subframes due to trees. I stacked them with DSS and did the usual post-processing in PixInsight but the noise was just too high. From watching a YouTube video I learned how to use a certain noise reducing filter in PixInsight (I can get you the name of it if interested). Today I opened up an account at AstroBin. When I uploaded the image I was surprised to see that Astrometry.net reported that the pixel scale was now 3.866 arcsec / pixel. After verification I came to believe that PixInsight is essentially doing 2x2 binning on the image in order to reduce noise!

Here is a link to the image:
http://www.astrobin.com/304948/?nc=user

Thanks,
Brian

User avatar
turfpit
Posts: 99
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: Binning ASI120MC

Post by turfpit » Fri Jul 28, 2017 7:15 am

Brian

That is a remarkable picture. I looked up the specs on the Travelscope 70. It looks like your substantial efforts with the stepper motor electronics and programming have been worth it.

Dave

User avatar
admin
Site Admin
Posts: 820
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Binning ASI120MC

Post by admin » Fri Jul 28, 2017 7:36 am

Hi Brian,

that is a great image for the equipment you are using and you're right that somewhere along the way it has been binned down from the ~1280x960 resolution of the ASI120 to half that (which of course will reduce the noise).

The trick with the approach you're taking (which is great fun and highly satisfying) is to always make sure you are focusing on the aspect that is holding you back rather than pushing one thing to the point of excellence when you still have problems elsewhere. Things that spring to mind from experience include the fact that the ASI120 gets very noisy at long exposures, so if you are planning on pushing subs beyond 120s you need to start thinking about cooling perhaps. While you are sticking with the Travelscope you will be limited in crispness of focus due to chromatic aberration, but for certain targets a narrowband filtered approach fixes that. So many things to play with :)

cheers,

Robin

User avatar
oopfan
Posts: 105
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Binning ASI120MC

Post by oopfan » Fri Jul 28, 2017 12:34 pm

Thank you Robin and Dave for your compliments and encouragement. As a programmer, and jack-of-all-trades-but-master-of-none as my Dad would say, I agree that the key to sustained progress is pulling only one lever at a time and finding the one with the best risk-to-reward ratio. I agree that my camera is getting awfully noisy but I am hoping that with the coming colder weather that it will be somewhat mitigated. There are a number of videos on YouTube where they've hacked my camera to add TEC cooling for very little money -- that pleases me. Also I can see a day when I break my cardinal rule and purchase a fast APO but then I will want a better camera! It is a vicious cycle. Just as long as I can continue hacking I will be pleased.

User avatar
oopfan
Posts: 105
Joined: Sat Jul 08, 2017 2:37 pm
Location: New York
Contact:

Re: Binning ASI120MC

Post by oopfan » Fri Jul 28, 2017 2:39 pm

I published a photo and description of my set up:

http://www.astrobin.com/304929/?nc=user

User avatar
admin
Site Admin
Posts: 820
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Binning ASI120MC

Post by admin » Fri Jul 28, 2017 4:03 pm

A nice simple setup put to great use :)

I cooled my ASI120MM at one point with a homemade cooler - component list:

Old CPU Heatsink/fan
£6 12V peltier from Ebay
Cable ties to hold above to camera
Bubblewrap and tape to insulate sides of camera.

You need a moderately powerful 12V PSU (3A+ or so). The effects are quite impressive in terms of reducing noise although the mono is more forgiving of noise in the first place (I think this is because a hot pixel in mono is just a single white dot, but on a colour sensor a hot red pixel affects many surrounding pixels when the debayering is applied to make the colour image).

cheers,

Robin

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest