#1  
Old 06-10-2017, 06:13 PM
Astrofriend's Avatar
Astrofriend (Lars)
Registered User

Astrofriend is offline
 
Join Date: Aug 2017
Location: Stockholm, Sweden
Posts: 439
Time efficiency simulations of three cameras

Hi,
When boring times comes it could be both fun and interesting to do some calculations. This time I did an Excel sheet over the three cameras: Atik 16200, Canon EOS 6D and ZWO ASI1600. I have the Canon 6D and wanted to see how time efficient it is compare to the others.

In the Excel sheet it's possible to setup data for the cameras and object information, light pollution and readout noise and max level that shouldn't saturate (clipp). At the end you see how many sub images it takes to reach a given S/N, Signal / Noise ratio. It also calculate the total exposure time and dead time between images. It only compare pixel to pixel, not that the sensor area or the pixel size is different.

Take a look here where I have wrote it down if you find it interesting:
http://astrofriend.eu/astronomy/tuto...fficiency.html

It's very simple so don't expect it to be perfect but you can have a lot of interesting information from it if you test with different parameters.

You can download the Excel sheet if you find it interesting.

/Lars
Reply With Quote
  #2  
Old 06-10-2017, 06:22 PM
LightningNZ's Avatar
LightningNZ (Cam)
Registered User

LightningNZ is offline
 
Join Date: Oct 2011
Location: Canberra
Posts: 951
Very interesting reading. Thanks
-Cam
Reply With Quote
  #3  
Old 06-10-2017, 06:31 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Interesting.

I would be shocked if the 6D could match your 16200 Atik in any way though.

I suspect the QE figures of the 6D as well. Some digital cameras like Sony have a funny way of expressing QE. Its unlikely that it is as high as you have stated in your calculations.

I have used high end OSC and whilst they are good that can't match a good mono CCD of which the 16200 is one of the current very best out there.

Greg.
Reply With Quote
  #4  
Old 06-10-2017, 07:26 PM
Astrofriend's Avatar
Astrofriend (Lars)
Registered User

Astrofriend is offline
 
Join Date: Aug 2017
Location: Stockholm, Sweden
Posts: 439
Thanks for the comments!

@Greg:
Me too, but that is the QE I found and it's not from Canon, it's the peak in the green spectra. My old Canon 5D (mark I) says to have QE=0.25 (green channel).

Always when see DSLR camera specs it looks too good. But both ZWO and QHY has CMOS cameras today with fullframe sensor from DSLR cameras (Sony). Just the electronics is different. When they come as a monocrome sensor I think they will be very interesting.

Now this test was good for Canon because I only calculated the green channel were it perform at its best. The IR cut filter block part of the red channel, but you can take it away if you want. The Canon DSLR isn't cooled either, but in Sweden when we do observation it's almost always below the frezzing point so it's not that important.

Christian Buil have done a lot of camera test, mostly on older Canon DSLR cameras:
http://www.astrosurf.com/buil/cameras.htm

I think I can trust him, he also find when comparing with a mono CCD that it performed much better. But you can't compare a green spectra with the whole visuable range. You must filter the monocrome camera too.

Maybe I can add more parameters later that do this calculation better.

You also have to take in mind that the Canons fullframe sensor take a much bigger area for every exposure, if you shall cover the same area with a smaller sensor and build mosaics it takes very long time.

Almost all DSLR cameras today only have 14 bit resolution, but the ASI1600 have even less, only 12-bit and it still performs very well what I have understand. When doing a lot of sub images you build up the dynamics from it as well as a 16-bit camera.

By the way, It could something wrong in my calculations also, this is the first version!


/Lars
Reply With Quote
  #5  
Old 06-10-2017, 07:59 PM
Atmos's Avatar
Atmos (Colin)
Ultimate Noob

Atmos is offline
 
Join Date: Aug 2011
Location: Melbourne
Posts: 6,980
The fact that the ASI1600 only has 12 bits isn’t that important. The 6D at its lowest ISO only has 11.38 bits of DR even if it does output 14-bit images. The ATIK 16200 sensor has 12.1 bit although the FLI16200 might be as high as 12.7 bits in a 16 bit file.

The ASI1600 is about 12.5 bits so you up the gain until you only have 12 bits. On my QHY163M that is at 80 gain but it is 75/76 on the ASI1600. What makes a camera like the QHY367M (Nikon D810) is that it has about 13.85 bits of DR! That’s why the D810 needs 14 bit files.
Reply With Quote
  #6  
Old 07-10-2017, 10:16 AM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Good reply Colin.

I have read in the past this sort of argument when most cameras were only 12 bit. As Colin said those types of cameras did not even use all the available data ability of 12 bit.

Not sure what 16 bit does in a modern CCD image except give the potential to show an additional level of brightness or darkness.

In CCD imaging the data is confined to this rather small band of the capability of 16bit. You have to use levels and curves several times to narrow it down to displaying the small range of data you captured. Most of that capability is thrown away.

One thing is for sure, modern digital cameras have some pretty hot performing sensors that just keep on getting better.

Sony seems to have several major innovations it has patented.

One likely one is RGBW (W=clear) sensor colour matrix. Its something like 60% brighter than a Bayer matrix sensor. Another is a curved sensor probably only useful in a fixed lens camera. Another is several versions of improvements on the Foveon style sensor where every pixel picks up RGB data. One has organic layers as well as silicone pixels.

So there are likely future large steps forward in sensor sensitivity that will benefit astrophotographers. It may well be they will challenge dedicated astro cams within 5 years or less at this rate.

Greg.
Reply With Quote
  #7  
Old 21-01-2018, 10:14 AM
Astrofriend's Avatar
Astrofriend (Lars)
Registered User

Astrofriend is offline
 
Join Date: Aug 2017
Location: Stockholm, Sweden
Posts: 439
Hi,
I find it very confusing how they measure the dynamic range. They compare a low ISO, say 200 where they reach the full well capacity and compare that with a high ISO, say 3200 where the readout nosie is at lowest.

Is this interesting to know?

When working in low nosie mode at ISO 3200 you have a very low roof.

If the gain at ISO 3200 is 4 (unitygain at ISO 800), and the Full Well capacity is 50000e-, then you just use the first electrones 0 to 4000 (14-bit ADC). The electrones from 4000 to 50000 you just trash away. With a 16-bit ADC you use the electrones 0 to 16000 with the gain 4. But normally there is much more complicated to construct a low noise 16-bit ADC if it shall be fast which are one of the demands in a DSLR.

If you have very low readoutnoise you can do a lot of short time exposures and get high dynamic, but the DSLR cameras mechanical shutter wont last for long.

/Lars
Reply With Quote
Reply

Bookmarks

Tags
ccd, cmos, dslr, qe efficiency, readout noise

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 11:22 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement