Re: [APML] Res, Info

[Matt BenDaniel <matt@starmatt.com>]

>> Matt wrote:
>> I use medium format PPF (e.g. at 875mm f.l.). 
>> Why does that improve my SNR over 35mm 
>> PPF (e.g. at 438mm f.l.)?
>
>Aplanatic@aol.com wrote:
>
>Hi Matt,
>What variables would you like to keep constant when discussing SNR?
> 
>Here are some of the possibilities:
> 
>Image scale in arcseconds/pixel
>Aperture
>Focal length
>Exposure time ("Sky Fog Limit")
> 
>In the end, if the detector is linear then, at the same image scale, the SNR depends only on the aperture
>and the exposure time. In your example, the SNR should be close to the same. However, if you expose each
>image to the same background density ("Sky fog limit") then, again at the same image scale, the downsized
>medium format image would have a higher SNR.
> 
>Dave Rowe

Dave,

Let me see if I understand.

Aperture multiplied by exposure time is proportional to the number of
photons entering the system.  My interpretation is that SNR has a lot to
do with number of photons collected.  The SNR obtainable with a
recording medium is also a factor of its quantum efficiency and dynamic
range.

Decreasing image scale (i.e. increasing focal length) increases SNR
because it deals with the detector resolution constraint (i.e. film
grain).  Increasing focal length only works up until seeing becomes the
constraint.

If I want to produce an image with a FOV of 4.6 degrees, I can use a
130mm aperture @ f/6.7 and expose on ISO 400 6x7 film for 120 minutes,
achieving a certain SNR.  [Assume the transparency and seeing are
excellent and the film has no reciprocity failure.]

To achieve the same SNR with 35mm ISO 400 film (same emulsion), I could:
a) Mosaick four shots on the same telescope.
b) Use a 0.5x reducer, yielding f/3.4.  I would only need to expose for
30 minutes to reach sky glow.  Stack four 30m shots.
c) Use a another telescope with a 65mm aperture @ f/6.7 and expose for
120 minutes.  Stack four 120m shots.

I am still confused, because options b & c have 2x the image scale.  For
b & c, should I scan the shots at twice the resolution (e.g. 4000 DPI)
before stacking?  Is that a form of drizzling?
-- 
Matt BenDaniel
http://starmatt.com

--  APML Archives at <http://astro.umsystem.edu/apml/>  ---
             Unsubscribe at <majordomo@seds.org>