Return to the Cloudy Nights Telescope Reviews home pageAstronomics discounts for Cloudy Nights members
· Get a Cloudy Nights T-Shirt · Submit a Review / Article

Click here if you are having trouble logging into the forums

Privacy Policy | Please read our Terms of Service | Signup and Troubleshooting FAQ | Problems? PM a Red or a Green Gu… uh, User

Announcements and News >> Discussion of CN Articles, Reviews, and Reports

Pages: 1 | 2 | (show all)
Craig
Vendor (Stark Labs)
*****

Reged: 09/16/03

Loc: Irvine, CA
Re: SNR: Image Sampling Myths - Part 5 new [Re: gdd]
      #3781358 - 05/02/10 12:46 PM

Quote:

Hi Craig,

I want to compare to setups resulting in the same number of photons delivered to each pixel on average. I am limited by the precision of my mount to a certain length subexposure for a given focal length. I can either buy a lens 1 stop faster or I can take twice as many subexposures. The resolution should be identical. Is the SNR lower for the slower lens? If so, is it because the subexposure is twice as long or because there are twice as many subexposures?





Gale,

Welcome to CN!

The spreadsheet posted with part 2 should actually let you run this kind of simulation. Exactly how the trade off works will depend a bit on your particulars. But, overall, you'll be better with the faster lens. You've got less dark current noise going into your stack and you've got less read noise (another way to think of this is that for a given sub-exposure length your faster f-ratio has higher SNR in each sub).

With 2x the photons going into a single sub the signal has gone up 2x. The SNR isn't up 2x as the noise part has gone up as well. Instead of sqrt(Target+Sky+Dark+read^2) we've got sqrt(2*Target+2*Sky+Dark+read^2). At worst, this will be 2/sqrt(2), aka sqrt(2), aka 1.414x better in terms of SNR. So, with no dark and read noise, we're at 1.414x the SNR. To make up for this with the slower lens, we need to take twice as many exposures.

Now, insert some dark current and some read noise and see what happens here... Try it! (Just fake some numbers for these and see what happens)

Craig


Post Extras: Print Post   Remind Me!   Notify Moderator  
Craig
Vendor (Stark Labs)
*****

Reged: 09/16/03

Loc: Irvine, CA
Re: SNR: Image Sampling Myths - Part 5 new [Re: Barry E.]
      #3781371 - 05/02/10 12:54 PM

Quote:

Craig,
Does this method of measuring the seeing work accurately with a DSLR since it has a Bayer matrix and an anti-aliasing filter? For example, the seeing is 2", but you are imaging at 2.5"/pixel. You won't be able to measure seeing below the image resolution will you?




The Bayer matrix will blur things a touch with a decent debayer but only a touch. The resolution you're imaging at is dictated by your pixel size and by the focal length, not by the fact that it's one-shot color or not. Of course, if you're trying to estimate a 3" FWHM and you're sampling at 30"/pixel, you'll never do it. Increase the focal length and you will.

But, you can do it with a OSC chip, just realizing that you'll over-estimate the blur by a bit. Have a look here at a demonstration of what is happening to your resolution:
http://www.stark-labs.com/craig/articles/assets/Debayering_API.pdf
http://helixgate.net/Xm%2DXCM%2D1A.html

You can also use a few tricks to get a mono image at full-res on a OSC camera. If you've got the raw data you can use something like Nebulosity's OSC Generic nebula filter reconstruction to balance the R, G, and B channels into a mono image without interpolation.

Craig


Post Extras: Print Post   Remind Me!   Notify Moderator  
Jared
Post Laureate
*****

Reged: 10/11/05

Loc: Piedmont, California, U.S.
Re: SNR: Image Sampling Myths - Part 5 new [Re: Craig]
      #4392584 - 02/17/11 01:14 PM

Craig, one thing you didn't cover in your essays (unless I missed it somewhere) is that it is possible to achieve a given SNR with a fixed aperture and exposure time by playing with one more variable--pixel size. If you are shooting with a telescope at f/9 with 9 micron pixels, that should be functionally equivalent to shooting with an f/5.4 scope of the same aperture using 5.4 micron pixels. Same spatial resolution, same number of photons per pixel per unit time. If the cameras have the same megapixel count you would even get the same field of view and therefor the same "information" content. Would you agree with this?

I guess the key is to balance spatial resolution against SNR keeping in mind the field of view desired. Obviously, you can't change pixel size at will (aside from binning, of course), but you can choose a camera that has a pixel size that is appropriate for your focal ratio, even if it is a fairly slow focal ratio.


Post Extras: Print Post   Remind Me!   Notify Moderator  
poita
member


Reged: 07/25/11

Re: SNR: Image Sampling Myths - Part 5 new [Re: Jared]
      #5200864 - 05/01/12 08:57 PM

So how does this apply to a system like the hyperstar that is imaging at F1.9 or F2 at around 500mm focal lengths?

Post Extras: Print Post   Remind Me!   Notify Moderator  
Pages: 1 | 2 | (show all)


Extra information
0 registered and 5 anonymous users are browsing this forum.

Moderator:  iceblaze 

Print Thread

Forum Permissions
      You cannot start new topics
      You cannot reply to topics
      HTML is disabled
      UBBCode is enabled


Thread views: 8476

Jump to

CN Forums Home


Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics