Re: Best Webcamera for Astroimaging currently?

On Sun, 15 Feb 2009 08:39:05 -0500, "Jason Banco" <a...@xxxxxxx> wrote:

You can say that, but I find a different conclusion. Also the fact that the
cooled cameras are 16 bit for the most part.

There is no such thing as a 16-bit camera. The best astrocameras are
nearly 14-bit, but most are 12-13 bits. That's a significant improvement
over webcams which are at most 10 bits (usually processed internally to
8 bits), but you only see that advantage if your telescope is large
enough, or your exposure time long enough, to fill those larger wells.
Doing so improves S/N against a readout noise floor that is about the
same in both cases. But in bad seeing, you normally will reduce your
exposure time as much as possible, which also reduces the signal-
usually to less than 8 bits unless you have a very large aperture scope,
irregardless of what your camera's maximum well depth is.

I guess I'm not making my point here, so I'll try again. Let's look at it
this way: I use a Toucam in poor to average seeing. I set for 1/15 sec
exposures and then capture 1000 frames of Saturn over and over again
throughout the night. After reviewing the frames, I find only 50 or so as
best quality and stack them. However, because the Toucam individual frame
quality is so poor, I can do no processing on these. A person then says,
well you didn't have good enough seeing. Why then can I do this process
manually, with a cooled 16 bit CCD camera, end up with 20 or so decent
frames, stack them, apply processing and the result is *always* better?

As I said before, my conclusion is that the problem is with your webcam.
An individual 1/15 second frame from a Toucan should be no worse than an
individual 1/15 second frame from any cooled, one-shot color
astrocamera. If you're seeing a difference, it's because your Toucam is
broken. And in bad seeing, you should be using shorter exposures than
1/15 second. That will really improve your end results, but of course,
you'll need enough frames to add up to a reasonable total exposure time.

Another possibility is that you have good seeing, and you're just
describing it as bad. With your long 1/15 second exposure, you may be
able to take advantage- in a single frame- of the greater dynamic range
of the astrocamera. That means each frame as somewhat better S/N. And
with good seeing, you may not need many frames to sort for quality.

Thousands of frames doesn't matter if the
quality is not there to begin with.

Thousands of frames is the only way to simultaneously beat seeing and
get good S/N. That's a fact based on physics alone, and has nothing to
do with the type of camera. Lucky imaging is called that for a reason.
The worse your seeing, the more frames you will need to get lucky.

Like I said, it does and I've seen this time and time again, which is why I
ditched webcams and stuck with the SX.

Well, you should use whatever produces results you like. But my advice
to the OP is to not take your path, because it makes no sense and is
contrary to anybody else's experience I'm aware of- including the other
imagers you have named in this thread.
_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com
.

Relevant Pages

  • Re: A very unique new camcorder !!
    ... highest frame rate and their highest resolution, ... MCT's camera, as an example, does 500 million frames per second ... high-speeds, but recording at 4k at high-speed, no camera can do this. ...
    (rec.video.production)
  • Re: Options for increasing source filter performance?
    ... I didn't mean to imply that the sleep was in FillBuffer() it's actually ... in the other worker thread which is queueing frames from the camera. ... In terms of getting a lower bit level image from the camera, ... Rendering the output pin ...
    (microsoft.public.win32.programmer.directx.video)
  • Re: A very unique new camcorder !!
    ... MCT's camera, as an example, does 500 million frames per second ... on YOUR NEWSGROUP. ... time making it to market. ...
    (rec.video.production)
  • Re: Interfacing Camera to Matlab
    ... 10 frames per trigger using the selected source. ... Log data to 'memory' on trigger. ... If the preview window is closed when you call getsnapshot, the camera is not acquiring data therefore the automatic gain control isn't working. ... Since it takes a few frames to get good images back, you'll always get black images that way. ...
    (comp.soft-sys.matlab)
  • Re: Milky Way from Death Valley
    ... To capture that much detail with an exposure time short enough to not show any star trails would take some very fast ... The photographer probably took this with a digital camera by finding ... Then the rest of the panorama frames were taken. ...
    (sci.astro.amateur)