Sign in with
Sign up | Sign in
Your question

ISO noise vs. long exposure noise

Last response: in Digital Camera
Share
Anonymous
August 27, 2005 12:40:02 AM

Archived from groups: rec.photo.digital (More info?)

Does anyone know of any data about ISO noise vs. long-exposure noise? In
other words, at what point does long-exposure noise become worse than
the noise you'd add by moving to a higher ISO? I've got a D70, and I'm
trying to take several-minute-long photos of dunes lit only by the moon,
but I'm not happy with the results I'm getting. (Someone's got to know
this - Roger Clark??)
Anonymous
August 27, 2005 12:40:03 AM

Archived from groups: rec.photo.digital (More info?)

Cynicor wrote:

> Does anyone know of any data about ISO noise vs. long-exposure noise? In
> other words, at what point does long-exposure noise become worse than
> the noise you'd add by moving to a higher ISO? I've got a D70, and I'm
> trying to take several-minute-long photos of dunes lit only by the moon,
> but I'm not happy with the results I'm getting. (Someone's got to know
> this - Roger Clark??)

Here is one study:
http://www.clarkvision.com/astro/canon-10d-signal-to-no...

Roger
Anonymous
August 27, 2005 1:43:47 AM

Archived from groups: rec.photo.digital (More info?)

On Fri, 26 Aug 2005 20:40:02 -0400, Cynicor
<j..tru.p.i.n...@speakeasy.net> wrote:

>Does anyone know of any data about ISO noise vs. long-exposure noise? In
>other words, at what point does long-exposure noise become worse than
>the noise you'd add by moving to a higher ISO? I've got a D70, and I'm
>trying to take several-minute-long photos of dunes lit only by the moon,
>but I'm not happy with the results I'm getting. (Someone's got to know
>this - Roger Clark??)

Just remember to use the noise reduction (Dark Frame Subtraction)
of the Nikon. This will remove "hot" pixels and amplifier noise
that generates "dark current" when longer exposures are used.
Also, you can chill your camera down to about 35 deg. F. which
dramatically reduces dark current. However, exposing the camera to
a much hotter photographic environment would probably fog the lenses.
But, if you take a picture now, for say 15 seconds, and then take one
when your camera has been outside on a winter day and has chilled, you
will see almost all dark current noise disappear. This is why
astronomers use dry ice and liquid nitrogen to chill CCDs used for
long exposures.
ISO noise is a different issue. Not much you can do about it
except noise reduction "out of camera."


"Bittorrents are REFUNDS for all the BAD movie products Hollywood
never gave us refunds for in the past"
Related resources
Anonymous
August 27, 2005 2:48:13 AM

Archived from groups: rec.photo.digital (More info?)

Roger N. Clark (change username to rnclark) wrote:
> Cynicor wrote:
>
>> Does anyone know of any data about ISO noise vs. long-exposure noise?
>> In other words, at what point does long-exposure noise become worse
>> than the noise you'd add by moving to a higher ISO? I've got a D70,
>> and I'm trying to take several-minute-long photos of dunes lit only by
>> the moon, but I'm not happy with the results I'm getting. (Someone's
>> got to know this - Roger Clark??)
>
>
> Here is one study:
> http://www.clarkvision.com/astro/canon-10d-signal-to-no...

I knew that if anyone had looked at it, it would be you. Thanks. :D 

Now the next question is why the guy with the new Rebel pointed at my
lens hood today and asked what it was, and as a follow-up, would he ever
get anything beyond his kit lens?
Anonymous
August 27, 2005 5:32:05 AM

Archived from groups: rec.photo.digital (More info?)

Tricky problem. You really need to understand the cause and effect in noise
theory for CCDs to answer this question. There are three choices, and all
offer slightly different results:

1) D70 at ISO 100 using a long exposure (turn the long-exposure noise
removal on and off for comparison)

2) D70 at ISO 1600 using short exposure (make sure it's really short, not
more than 1/2 otherwise you are measuring a subtle combination of things)

3) Several images using D70 at ISO 400 and "medium" exposure time. Add
these images via post processing together. This is also known as manual
integration. Can work great.

To learn about digital noise and its source, use Google. But much has been
said and I recommend that your try the three scenarios with different
settings until you find what works best for you. Your own data will speak a
thousand words!

For your application, I would try #2 and #3 since an ISO of 1600 will most
likely not give you exposure times shorter than 1/2 seconds.

Also remember, there are very good digital noise removal filters available
for many programs (e.g. Nikon Capture 4.3.1).

Please let us know your results.

Gregor



"Cynicor" <j..tru.p.i.n...@speakeasy.net> wrote in message
news:rMCdnYv828_5K5LeRVn-1Q@speakeasy.net...
> Does anyone know of any data about ISO noise vs. long-exposure noise? In
> other words, at what point does long-exposure noise become worse than the
> noise you'd add by moving to a higher ISO? I've got a D70, and I'm trying
> to take several-minute-long photos of dunes lit only by the moon, but I'm
> not happy with the results I'm getting. (Someone's got to know this -
> Roger Clark??)
Anonymous
August 27, 2005 10:14:03 PM

Archived from groups: rec.photo.digital (More info?)

"Cynicor" <j..tru.p.i.n...@speakeasy.net> wrote in message
news:rMCdnYv828_5K5LeRVn-1Q@speakeasy.net...
> Does anyone know of any data about ISO noise vs. long-exposure noise? In
> other words, at what point does long-exposure noise become worse than
> the noise you'd add by moving to a higher ISO? I've got a D70, and I'm
> trying to take several-minute-long photos of dunes lit only by the moon,
> but I'm not happy with the results I'm getting. (Someone's got to know
> this - Roger Clark??)

Here is a shot from a 20d for 1153 seconds (20 mins er so).
http://harryphotos.com/trucklogo.jpg
full moonlight, noise reduction mode off, iso 100, I was very impressed. I
think the D70 uses a ccd which can heat up with longer exposures. Something
else you can try is taking a bunch of shorter pictures and stacking them.
Here is 40x30 second exposures stacked: http://harryphotos.com/bridge2.jpg
there are slight gaps in the stars however (I was using mirror lockup and
the 2 second delay was long enough, I would suggest not using it, noise
reduction mode is out of the question if you want smooth star trails.)
!