Film or Digital Camera

I've been arguing with a friend who has a telescope about using film
instead of a digital camera. He insists that film is not as good as
using the digital camera and some stacking software. I always believed
that film is the ultimate for long exposures because there's no noise
polluting the image. I could be mistaken, though since I haven't had
any astro-photography experience, (save 20 digital photos).

The principle question I have for the experts (and I don't mind
finding out I was wrong), is: Will film outperform a digital camera
when doing long exposures?

So far, my longest exposure was 3 minutes and I saw low level noise
when I used photoshop(adjust levels) to brighten up the frame. I
haven't tried 10 minute exposures yet but I would like to try film for
10 minutes. Is film worth the inconvenience of not seeing the image
right away?

Thanks for any advice.

I forgot to add that the digital resolution is 3000 X 2000 so I don't
think it's an issue in this context.

On Sun, 27 Jul 2003 20:07:54 GMT, Dave J. wrote:

I've been arguing with a friend who has a telescope about using film
instead of a digital camera. He insists that film is not as good as
using the digital camera and some stacking software. I always believed
that film is the ultimate for long exposures because there's no noise
polluting the image. I could be mistaken, though since I haven't had
any astro-photography experience, (save 20 digital photos).

The principle question I have for the experts (and I don't mind
finding out I was wrong), is: Will film outperform a digital camera
when doing long exposures?

If by digital camera you mean an uncooled camera designed for normal terrestrial
imaging, then in general film will outperform it for most deep sky imaging. Film
is thus far capable of much longer exposures than these types of digital cameras
and can therefore reach deeper, in spite of its inherent lower sensitivity.
There are new digital cameras with low noise, although this noise remains high
compared to cooled cameras. But you are mistaken that there is no noise when you
use film. In particular, sky fogging brings the background up and reduces the
S/N, and there are various non-linearities. Also, most films are lower
resolution than the sensors in most digital cameras.

Cooled digital cameras are vastly superior to film.

_________________________________________________

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

"Dave J." wrote in message
...
I forgot to add that the digital resolution is 3000 X 2000 so I don't
think it's an issue in this context.
Funnily enough, it probably is.
This makes it allmost certain, that this is a CMOS sensor (a CCD this size
would be very expensive). Depending on the manufacturer, you may well be
able to get some quite 'acceptable' images, but the results will not be as
noise free as CCD images. The sensitivity will still be better than film,
but not up to the class of properly designed astronomical cameras.

Best Wishes

The gurus here will surely point out the many attributes of the various
mediums, so rather than beat that drum, I'd like simply to make the
following observation: digital camera astro-imaging with good-to-great
results can be as simple as falling off a log... period!

I've not seen any notable film-based or dedicated astro-CCD images that were
not the result of a lot of (Ahem...pardon the acronym!) "Special Highly
Intensified Training" invested by those taking the shots. Perhaps I'm
mistaken, and someone has indeed just walked up to a scope, plonked the
equipment into place, and began firing off round after round of winning
images!

My own feeble attempts with film-based astro-imaging left quite a bit to be
desired. Despite my best efforts and seemingly innumerable calculations,
even my best efforts looked only vaguely like what I had anticipated. After
more than a year of this tortuous routine, I decided to try using a digital
camera (insert blinding light and the fanfare of trumpets here), and the
difference was phenomenal! I could now see what I was capturing right then
and there, and if I successfully bagged it or needed to make adjustments. No
big learning curve (though techniques obviously can be improved upon with
experience), no major disappointments from the processing lab, and no
multi-thou$and dollar outlay for dedicated equipment (although you could
spend many thousands, you don't need to to have fun with digital camera
astro-imaging).

Bottom line:

Film can be very frustrating.. even for those who have considerable
experience working with it.

Dedicated astro-CCDs require a learning curve, usually a bit of $$$, and
various other gear (laptops etc)

Your aunt Matilda's digital camera (the one she brought to the family
picnic) can provide surprisingly good astro-images with relatively little
effort.

Finally, if something is too much work, too epen$sive, or just too darned
frustrating, what do you think the chances are that you will enjoy it?!? )

Paul

--- http://www.astro-nut.com/ ---


"Dave J." wrote in message
...
I forgot to add that the digital resolution is 3000 X 2000 so I don't
think it's an issue in this context.

One factor in favor of digital is that the images can be immediately
processed on a
computer and enhanced to bring out the best the image has to offer. And
this can be
done the minute you get home. And it can be done without a darkroom or film
processing equipment.

Of course, you can take photos, have them transferred into digital images,
and then
manipulate them, but then you're at the mercy of how the developer attempted
to
develop and print the film (also scanning the print if you allow him to do
so).

And you need to develop the film in order to find out if you took any photos
good
enough for further processing.

Film still has one big advantage, and that is fine grain that no CCD can
match yet.
But it isn't as sensitive as CCDs, and doing your own darkroom processing
still
requires much more room and equipment than digitial processing.

Just my $0.02 worth.

Clear and Steady Nights !
--- Dave

On Sun, 27 Jul 2003 22:19:06 GMT, "Paul Hyndman"
wrote:

Bottom line:

Film can be very frustrating.. even for those who have considerable
experience working with it.

Dedicated astro-CCDs require a learning curve, usually a bit of $$$, and
various other gear (laptops etc)

Your aunt Matilda's digital camera (the one she brought to the family
picnic) can provide surprisingly good astro-images with relatively little
effort.

Finally, if something is too much work, too epen$sive, or just too darned
frustrating, what do you think the chances are that you will enjoy it?!? )

The other thing that mustn't be overlooked is the fact that working in
the digital realm is also a whole lot of fun! Instant feedback is
quite addictive.

On the quality side, it's quite eye-opening, now that Mars is upon us,
to look back at some of the old film photos of Mars that were taken
from the mid 70's onwards (any 'old' general astronomy book will have
one). A lot of these images were taken with big scopes. A lot of
them are *very* inferior to the modern digitally enhanced results
taken by amateurs today.

--
Pete
Homepage at http://www.pbl33.co.uk
CCD/digicam astronomy

On Sun, 27 Jul 2003 21:48:04 +0100, "Roger Hamlett"
wrote:

... Even with an 'uncooled'
chip, you can massively reduce the noise, by taking a series of 'dark
frames', combining these (median combine), and subtracting this from an
image of the same duration.

Sorry if I'm being pedantic, but this is often stated and entirely incorrect.
Noise can never be reduced, even a tiny bit, by subtraction. There are two
primary systemic noise sources, dark current noise and readout noise. The first
is reduced by lowering the chip temperature, the second is fixed. The effects of
readout noise can be reduced by reducing the number of images that are stacked,
and by increasing the length of individual exposures. Subtracting a dark frame
merely removes a bias level, leaving behind exactly the noise that was already
there. This can be done with film, also, although the non-linear response of
film makes it a bit trickier.

There is also statistical noise caused by the uneven rate that photons are
collected. This is minimized only by collecting more photons.

_________________________________________________

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

I appreciate all the responses and I have the overall impression that
the digital camera I have now might be ok to stick with for the long
exposures. I forgot to say the type of camera. It's a Nikon D100.
There's a "concentration of noise" in the upper left corner of the
longer exposures. I think it might be heat on the chip or heat from
something in the camera. So I wish I could find out more about the
method used to stack images. I've seen a lot of discussions in USENET
but I don't have anything authoritative to read, just lots of
opinions. Is there a consensus on the best way to stack them? If
there's somewhere on the Internet I can read about how stacking works,
that would be great. Is there a "Stacking FAQ" somewhere?

Thanks

On Mon, 28 Jul 2003 05:40:37 GMT, Dave J. wrote:

I appreciate all the responses and I have the overall impression that
the digital camera I have now might be ok to stick with for the long
exposures. I forgot to say the type of camera. It's a Nikon D100.
There's a "concentration of noise" in the upper left corner of the
longer exposures. I think it might be heat on the chip or heat from
something in the camera. So I wish I could find out more about the
method used to stack images. I've seen a lot of discussions in USENET
but I don't have anything authoritative to read, just lots of
opinions. Is there a consensus on the best way to stack them? If
there's somewhere on the Internet I can read about how stacking works,
that would be great. Is there a "Stacking FAQ" somewhere?

Not that I'm aware of. But most image processing software allows you to combine
images. This can be done my adding or averaging, and in most cases there is
little difference. Usually, you will also need to perform some sort of alignment
between the images. This is very easy if you have stars, but a little more
difficult with planetary images. Still, there is a variety of software to
perform alignment, or partly automate it.

Particularly with an uncooled camera, you need to find the balance between dark
current noise and readout noise to identify the best exposure time to use. High
dark current (which results in high dark current noise) is minimized by
combining as many exposures as possible. High readout noise is minimized by
combining fewer exposures of longer length. I'm not sure about the D100, but
most digital cameras seem to image best at around 15-30 seconds, with some of
the newer ones giving good results out to several minutes. In any case, you will
want to acquire dark frames and subtract them from the individual images before
stacking. Some cameras will do this automatically, but I'd advise turning this
feature off and making manual darks. That's because the darks have noise, too,
and the way to reduce it is to combine many individual darks into a low-noise
master.

_________________________________________________

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