Starfields from "bulb" and RAW?

Good day, and thanks in advance for all help.

Can someone either confirm or refute an asumption that a digital camera
(7MP and 8MP) can capture starlight in a widefield shot using the camera's
"bulb" shutter setting and RAW image format?

Assume the camera is tripod mounted, and the camera's timer is used to
avoid vibration of pressing the shutter release.

Utilizing this method, I was able to get some decent shots of a moon/Venus
conjunction using an old 3MP camera. While stars (and even their colors)
were visible on the LCD viewscreen during attempted widefield starfield
shots, stars would not appear in a maximum 4-second exposure JPG image.

So, I'd like to know if higher mega-pixel resolution along with longer
exposures and uncompressed imaging would achieve the desired ends.

Thanks again!

NightGuy wrote:

Good day, and thanks in advance for all help.

Can someone either confirm or refute an asumption that a digital camera
(7MP and 8MP) can capture starlight in a widefield shot using the camera's
"bulb" shutter setting and RAW image format?

Assume the camera is tripod mounted, and the camera's timer is used to
avoid vibration of pressing the shutter release.

Utilizing this method, I was able to get some decent shots of a moon/Venus
conjunction using an old 3MP camera. While stars (and even their colors)
were visible on the LCD viewscreen during attempted widefield starfield
shots, stars would not appear in a maximum 4-second exposure JPG image.

So, I'd like to know if higher mega-pixel resolution along with longer
exposures and uncompressed imaging would achieve the desired ends.

The truth of the matter is that if you go longer than about 4 seconds,
you'll end up with star trails. Whether this is what you're looking for
or not is something you don't specify. You can, for instance, stack a
number of 4-second exposures to get a whole star trail exposure, with
some obvious caveats (multiplied noise floor, gaps in star trails
representing the moments between exposures).

Bulb mode makes this easier to accomplish, since you have an arbitrarily
long exposure. Shooting RAW gives you wider lattitude and the
opportunity for a sharper image. Image resolution doesn't have that much
to do with it, though of course the more pixels the better. :-)

Some hints: http://www.astropix.com/HTML/I_ASTROP/TRIPOD/TRIPOD2.HTM

If you want the stars without their trails, you'll have to come up with
a rig that compensates for the movement of the earth relative to the
stars.

--
http://www.xoverboard.com/cartoons/2..._argument.html

On Fri, 05 Oct 2007 22:40:32 GMT, (NightGuy) wrote:


Good day, and thanks in advance for all help.

Can someone either confirm or refute an asumption that a digital camera
(7MP and 8MP) can capture starlight in a widefield shot using the camera's
"bulb" shutter setting and RAW image format?

Assume the camera is tripod mounted, and the camera's timer is used to
avoid vibration of pressing the shutter release.

Utilizing this method, I was able to get some decent shots of a moon/Venus
conjunction using an old 3MP camera. While stars (and even their colors)
were visible on the LCD viewscreen during attempted widefield starfield
shots, stars would not appear in a maximum 4-second exposure JPG image.

So, I'd like to know if higher mega-pixel resolution along with longer
exposures and uncompressed imaging would achieve the desired ends.

Thanks again!

I have easily obtained wonderful images of the full path of the Milky-Way using
just a Sony F717 P&S camera. These P&S cameras are able to take images of stars
down to magnitude 7.5 (a few people have claimed 8.2), 7.5 is one full magnitude
less than is visible to the unaided eye on a very dark clear night away from all
light pollution. I selected some of the dimmer stars out of the images then
compared them to star charts to see how faint those particular stars were. I
have one really nice photo of Orion (with its Great Nebula of) and surrounding
constellations perfectly reflected in the water of a still-as-glass lake.

They do exceptionally well on aurora too. The wavelength of light put out by
excited nitrogen and oxygen atoms are just the right wavelengths to be detected
by these sensors. When all I can see is a very dim barely detectable gray patch
in the sky, even after my eyes have been dark adapted for 30 minutes or more,
the camera easily captures this as an intense green structure full of detail. I
never hesitate to try to get an aurora photo if all I can see is just a "that
might be, is it?" patch in the sky.

With a 15 to 30 second exposure you may get some star trails, but that depends
on the FOV of the lens. Wider angle allows for more motion while still keeping
your star images at near point-sources of light. You can also stack multiple
shorter exposures to get the same effect for zoomed-in shots, matching star to
star in each frame.

Putting a fish-eye adapter on these cameras is interesting. I can get a
whole-sky view of all the constellations. You lose a bit of magnitude reach from
the extra glass robbing some light, but it's nice to see all the constellations
at once in a photo that you recorded. Really nice for recording meteor showers
when you don't know what part of the sky they might fly though. Their radiant
location the direction from which they appear to travel but they can appear
anywhere in the sky from horizon to horizon.

Here's one of my test samples of what can be done with a decent P&S camera. The
tall grasses being lit by a street-lamp about a block away.

http://farm3.static.flickr.com/2355/...23cb3a48_o.jpg

The raw format saves the data in the high bit depth (usually 12 bits)
that comes out of the analog - digital converter as linear data.
This is extremely important for astronomical images.

Best Regards
Marcin Gorgolewski
www.gorgolewski.com

P.S. Don't forget to lock the mirror up before the exposure to reduce camera
movement.

On Sat, 6 Oct 2007 02:19:48 +0200, "Marcin"
wrote:

The raw format saves the data in the high bit depth (usually 12 bits)
that comes out of the analog - digital converter as linear data.
This is extremely important for astronomical images.

P.S. Don't forget to lock the mirror up before the exposure to reduce camera
movement.


I never have that problem. I use a P&S camera that's not so limited and fussy.

You don't need RAW, you don't even need Bulb mode. Just get a decent camera.
Mine happen to be P&S cameras that do all this. One doesn't even have RAW
capability and it takes some of the best star photos.

On Sat, 06 Oct 2007 00:19:34 GMT, Jim C. wrote:


Here's one of my test samples of what can be done with a decent P&S camera. The
tall grasses being lit by a street-lamp about a block away.

http://farm3.static.flickr.com/2355/...23cb3a48_o.jpg

In case you are not constellation savvy, I forgot to mention that that's Orion
in the right-of-center to upper-right of this photo. Taking up the right 3rd of
the image. Difficult to pick him out of all the other stars that aren't normally
visible to the human eye on most nights from most locations. You can pick Orion
out of the busy background of dimmer stars by looking for his 3-star belt
(almost perfectly horizontal on the right), his star+nebula+star sword hanging
from his belt (hanging in a 7-o'clock direction from the 3-star belt), and the
brightest star of Sirius in Canis Major (in the left of center above the
grasses). Due north is in about a 1-o'clock direction.

Most people can easily see the obvious shape of Orion because light pollution or
the limits of their eyes automatically drown out all the dimmer stars, except
for those that make up Orion. Not so in this photo. He gets lost in so many
dimmer star being recorded.

Paul Mitchum wrote:

Some hints: http://www.astropix.com/HTML/I_ASTROP/TRIPOD/TRIPOD2.HTM

If you want the stars without their trails, you'll have to come up with
a rig that compensates for the movement of the earth relative to the
stars.


Could one take multiple short frames, and do an aligned stack
to compensate (over modererate time spans) ?

BugBear

On Sat, 06 Oct 2007 12:24:52 +0100, bugbear
wrote:

Paul Mitchum wrote:

Some hints: http://www.astropix.com/HTML/I_ASTROP/TRIPOD/TRIPOD2.HTM

If you want the stars without their trails, you'll have to come up with
a rig that compensates for the movement of the earth relative to the
stars.


Could one take multiple short frames, and do an aligned stack
to compensate (over modererate time spans) ?

BugBear

This is a normal function of frame-stacking used in all astrophotography. Here's
the overview info from Registax (freeware) page.

http://www.astronomie.be/registax/html/v4_site.html

Note the 2 different alignment options:


Batch-mode Operation
RegiStax now can be operated also in a batchmode on AVI-files using a (currently
limited) set of operators that allow you to align/optimize/stack and save the
result. The command-list allows you to set:
-filename to be processed
-default directory
-alignmentpoint X,Y (for single alignment)
-alignmentpointfile (for multiple alignment)
-first/last frame to process (block processing)
-number of frames to be stacked
-create reference (number of frames)
-waveletsettings (filename)
-autosave results (format)
-autoexit after processing

Added Input formats:
Two sequential (multiple frames in a single file ) formats have been added both
aiming at the output of Lumenera stylecamera’s:
-16 bit (per channel) AVI format based on K3CCDtools format (Peter Katreniak)
-16 bit (per channel) SER format based on LucamRecorder (Heiko Wilkens)

Alignmentpage
- 3 alignmentmethods (single, multi-point and none)
- view full image (during alignment and during positioning of alignmentpoints)
- user defined alignmentbox (draw an alignmentbox -single alignment around a
feature)
- option to “ignore” warnings (and frames) that are misaligned
- allows you to define which messages (during alignment/optimisation) to show
- predefined fits/tif image-stretching (usefull when datasets for instance do
not span the full 16bits of a file makeing the preview look very dark)
- automated detection of actual used “bitdepth” of imageformats (only for
8bits) to the nearest 2^n level

On Sat, 06 Oct 2007 08:55:09 -0500, Jim Townsend wrote:

NightGuy wrote:

Good day, and thanks in advance for all help.

Can someone either confirm or refute an asumption that a digital camera
(7MP and 8MP) can capture starlight in a widefield shot using the camera's
"bulb" shutter setting and RAW image format?

Yes it's possible. The link below is a shot I made of the Andromeda
galaxy using a Canon 10D and a 50mm lens. (12 seconds, f/1.8, ISO 1600)

The galaxy is that horizontal smudge at the bottom right hand third
There's not that much detail :-)

http://www.mts.net/~jwt/images/andromeda.jpg

Interesting photo but I'm surprised at how much noise is in it. Quite a lot in
fact. I was under the impression that the sensors in those camera could do
noiseless high-ISO photography, especially for so short of an exposure. The
Milky-Way doesn't pass through that part of the sky. The outer edges of the
Milky Way are toward the top left of your photo but not passing through where
the galaxy lies. According to the orientation of the galaxy the stars of the
Milky-Way should be getting stronger to the top left of the image where yours
appears darker and dimmer to the bottom where your star-field gets lighter.
Mostly all of those dimmer stars are is just sensor noise, probably due to some
sky-glow enhanced by vignetting.

In cases like this it's good to do a histogram adjustment to clip out the base
noise that you know not to be stars. Using noise-reduction software is useless
because it invariably filters out valid stars. A good set of star catalogs
installed into Cartes du Ciel, like the USNO SA2 catalog for magnitudes down to
19 or 22 can help for extreme photography. But something like the Sky2000
catalog with stars down to magnitude 9, or Tycho 2 to magnitude 12 at the most,
should suffice for all single-frame digital camera photography. The center of
M31, the Andromeda Galaxy, is magnitude 3.4 for comparison. I did manage to find
a magnitude 9.6 star in your photo near the galaxy that I'm almost sure isn't
noise, but beyond that all that gray mottling is noise. Try to get the
background of the central part of the image as dark as the vignetting in the
upper two corners by clipping with a histogram tool and it should be just about
right. It might help to correct it with a gradient vignette-mask first to dim
the center or lighten the corners so when it's all clipped it's done evenly.

If anyone used that for trying to find something dim with their telescope they'd
be wondering where all the real stars went to.

shane_steiner wrote:
On Sat, 06 Oct 2007 08:55:09 -0500, Jim Townsend wrote:

NightGuy wrote:

Good day, and thanks in advance for all help.

Can someone either confirm or refute an asumption that a digital camera
(7MP and 8MP) can capture starlight in a widefield shot using the camera's
"bulb" shutter setting and RAW image format?
Yes it's possible. The link below is a shot I made of the Andromeda
galaxy using a Canon 10D and a 50mm lens. (12 seconds, f/1.8, ISO 1600)

The galaxy is that horizontal smudge at the bottom right hand third
There's not that much detail :-)

http://www.mts.net/~jwt/images/andromeda.jpg

Interesting photo but I'm surprised at how much noise is in it. Quite a lot in
fact. I was under the impression that the sensors in those camera could do
noiseless high-ISO photography, especially for so short of an exposure.

It is only a 12-second exposure! Probably boosted a lot from the original.
Plus the 10D is a lot noisier than modern cameras (Yes I own one and
have determined its quantum efficiency).
A 10D on a magnitude 9.6 star in a 12-second exposure with a
50mm f/1.8 lens collects a total of about 600 photons (most likely
spread over several pixels) if the star is directly overhead
(less if not).

An example of an M31 image with 28.75 minutes exposure and an
f/5.6 lens at ISO 800:
http://www.clarkvision.com/galleries...-v1.6-700.html

For the OP: It may be that your camera in long exposures does noise
reduction that deletes the stars confusing stars with noise.
Even Nikon DSLRs in raw mode do this! If your camera does this,
you probably need another camera unless you can turn off the
noise reduction.

If you are interested in night and low light photography, here is an
article that shows waht can be done:
http://www.clarkvision.com/photoinfo...ht.photography

For low light work, you need a camera with large pixels. See:

http://www.clarkvision.com/imagedeta...el.size.matter

http://www.clarkvision.com/imagedeta...rmance.summary

Roger