| Author | Thread |
|
|
11/23/2004 12:07:03 PM · #1 |
I have a (I hope) simple question. At high ISO numbers (800 or 1600) does digital noise get worse the longer the shutter is open? Or is it more of a function of the ISO alone at these high ISO numbers, and what you get you are going to get no matter the shutter speed? Anybody know? Thanks.
Doug
|
|
|
|
11/23/2004 12:14:55 PM · #2 |
| Both cause noise. But you will get more noise the longer the shutter is open. My g6 does an automatic black frame extraction if the shutter is open for more than 1 second, no matter which iso I choose. I love that feature. |
|
|
|
11/23/2004 12:20:18 PM · #3 |
Thanks. That's what I was thinking too, but wasn't sure...
Doug
|
|
|
|
11/23/2004 12:26:43 PM · #4 |
As usual, the question is really not that simple. The random noise level should not be dependent on shutter speed, however the fixed pattern noise will be. Both the fixed-pattern and random noise levels are increased with high ISO. Since dark frame subtraction compesates for fixed-pattern noise, it is useful for long shutter speeds, especially at high ISO. It is not useful in reducing random noise.
For reduction of random noise, use multiple exposures and average them later in software. Random noise is reduced according to the 1/(Square root) of the number of frames, i.e. 2 frames will decrease noise to 0.71x original, 4 frames will reduce to 0.5x, 8 frames by 0.33x, etc.
|
|
|
|
11/23/2004 03:16:15 PM · #5 |
I thought random noise (also called dark noise) would increase with longer shutter speeds. It is caused by thermal agitation generating electrons in pixels (like light does but not as quickly), and keeping the shutter open (and the sensor active) for a long time gives more opportunity for this to happen. But a high ISO amplifies the noise along with the signal. So given Exposure A with a slow shutter speed and low ISO and equivalent Exposure B with a faster shutter speed and high ISO, A will have more noise than B, but the noise in B will be a lot brighter than A.
If I have a choice (i.e., the subject is still and I have a tripod so a long exposure time is feasible), I personally prefer A since the noise is usually less obtrusive. But a high ISO does often allow handheld shots or photos of moving subjects in dark settings that wouldn't otherwise be possible, and that's better than nothing.
Note that a situation requiring both slow shutter speed and high ISO is probably so dark that the noise overcomes the signal. One way around this is, as kirbic mentions, to use multiple exposures and average them during postprocessing. (If your camera doesn't have a dark frame subtraction feature, take another exposure or two using the same settings with the lens cap on and subtract it yourself.) |
|
|
|
11/23/2004 03:18:33 PM · #6 |
| In my simple way, that is what I was trying to say, dr rick. Thanks. |
|
|
|
11/23/2004 04:00:39 PM · #7 |
I tend to get too technical. There's nothing wrong with being simple! Let me practice:
You get more noise the longer the shutter is open (like pcody said).
The higher your ISO setting the brighter the noise (the tradeoff). |
|
|
|
11/23/2004 04:10:03 PM · #8 |
I guess I should qualify my earlier post. I was thinking of a "given situation", that is, a single specific exposure situation. If you leave the shutter open longer, yes, there will be more opportunity for thermal noise, however the signal from incoming light will also be increased, so the signal-to-noise ratio will not decrease (so apparent noise level will not increase). If you're talking about purely the amount of random noise, without any signal, then yes, it will increase with increasing shutter time.
What's really important for image quality is the signal-to-noise ratio. Just as with any amplification task, as the signal (in this case the amount of light) gets smaller, the task of amplifying without noise becomes much more difficult. In really low-light situations such as astrophotography, it's pretty much an absolute necessity to do both averaging and dark frame subtraction. Cameras made specifically for astrophotography normally are cooled using Peltier (thermoelectric) coolers as well.
|
|
|
|
11/23/2004 06:49:22 PM · #9 |
Oh. I wasn't nagging at you dr rick. I thought your answer was much better than mine. It was a compliment that you could say exactly what I wasn't able to say.
And everything kirbic says is true. I get dizzy when I have to think of it though. I've taken physics and all those brain scrambling classes, but it still makes me want to upchuck when I have to force myself to think in that type of pattern. Yes/no answers in these cases are much more comforting. |
|
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 09/16/2025 11:53:15 PM EDT.
