Raw efficiency of the Mono Atik 383L+ vs 460EX for the same FOV ?

[Hibou]

I know there have been various comparisons of these cameras, so I apologise for raising it again, but my application is a little special :-)

I want the most sensitive way of imaging a fixed FOV (non-astronomical, where I can adjust the object distance to fit the CCD). I would bin x2 or x3 to get effectively bigger pixels since my "stars" are not sharp, and I am not too worried about the extra noise of the 383L+ even though exposures would be 1-5 minutes.

I reason that the the greater area of the 383L+ chip and the fact that it is full frame will mean that it will be brighter under these conditions, even though the 460EX chip has higher quantum efficiency and microlenses will help compensate for the interline area lost. The 383L+ is also cheaper, but my base comparison is the 414EX which is cheaper again but with a much smaller chip.

I would be grateful if anyone can find a flaw in this simple reasoning, or better some actual experience of both monochrome cameras under similar conditions ?

[vlaiv]

Maybe this will help. I did analysis for astronomical use but it was for same focal length (therefore different resolution, but by a small factor). I concluded that 460 is better choice. Reasons being:

460 QE : 383+ QE = 1.4

383+ pixel surface : 460 pixel surface = 1.5

So far 383+ is in the lead just by a bit, but for astronomical purpose and SNR, read noise and dark current are also important, and this is where 460 wins (5e read noise vs 9e read noise, also dark current is much smaller in 460).

Although being more expensive (this is important consideration in my case), versatility of 460 for my intended setup (pixel size binned/unbinned and sensor size in light of telescope aberrations and use of accessories such as filters and focal reducers) and better overall estimated efficiency proved decisive factors over price advantage and ccd surface of 383+.

In your case, well depends what your case is, if you are going to fix fov over entire surface of chip 383+ might have advantage being much larger chip, therefore you can do more aggressive binning and increase SNR that way. Don't know what resolution you are after, and what kind of low light level application you are doing. Also to get fixed fov over entire surface you will probably need different optics in front of ccd (or can this be done by distance to target?), and it's efficiency also plays a part I guess.

If you give a bit more details on what you are after, maybe we can conjure some kind of similar reasoning as I did for my astronomy use.

 

[Hibou]

1 hour ago, vlaiv said:

460 QE : 383+ QE = 1.4

383+ pixel surface : 460 pixel surface = 1.5

If you give a bit more details on what you are after, maybe we can conjure some kind of similar reasoning as I did for my astronomy use.

Many thanks vlaiv. I agree that if you consider the QE and pixel area they come out more or less equal. But I still expect the bigger chip to win because I can use the same f/1.4 optics to get the same FOV simply by adjusting the distance to my 120x90mm flat object screen. I might then use x4 binning with the 383L+ rather than x3 binning with the 460EX to get even larger pixels. And in principle a full frame CCD should also beat an interline CCD for sensitive area (the microlenses probably don't work quite so well at f/1.4 and high incident angles). OK, there may not be a huge difference, but perhaps still worth having especially since it is cheaper.

I am not so concerned by the higher dark current and read noise because I already use the Atik 4000 and it's not as bad compared to Sony chipped cameras as it might appear on paper (for my application :-). The cameras with Kodak chips also have better cooling.

Below is an example image, collected in 5 minutes with a pair of 314L+ cameras stitched together, a complication I would like to avoid. (I am flipping between the original image and an image massaged by imageJ to bring out faint spots; I am only interested in their positions). You can see why I am interested in astronomy cameras even though my application is quite different.