| Author | Thread |
|
|
07/07/2014 01:13:06 PM · #1 |
| In terms of noise, is it better to use a lower ISO and then push the exposure in post process, or use a higher ISO to get the proper exposure in camera? At what point does it become more sensible to use just a higher ISO? Can this be turned into a repeatable or testable "rule" to take out the guesswork? |
|
|
|
07/07/2014 01:33:53 PM · #2 |
In general, it is better to use the higher ISO. There may actually not be a big difference in the amount of noise between the two scenarios, but remember that you lose half your bit depth by underexposing by one stop. In other words, by "pushing" in post, you will end up stretching the data and winding up with very coarsely distributed values at the dark end of the scale.
That being said, as good as cameras are these days it is very viable to pull up a high ISO exposure by one stop and have completely acceptable results where even 5 years ago we would have had a much less successful outcome.
|
|
|
|
07/07/2014 01:39:14 PM · #3 |
that's a great answer. i forgot about the disadvantages of underexposure in terms of tone levels.
but i did hear that it was a bit of a myth that you lose half your tone values by underexposing in this way (or at least when worded in this way; maybe "bit depth" is more accurate). Could you clarify further about exactly what this means? |
|
|
|
07/07/2014 02:32:31 PM · #4 |
| I've been pushing my histogram more to the right lately, and it has a helped a ton in the effectiveness of my photos. |
|
|
|
07/07/2014 03:05:26 PM · #5 |
Here's a good discussion:
//photo.stackexchange.com/questions/35136/is-it-better-to-shoot-with-a-higher-iso-or-use-lower-iso-and-raise-the-exposure
I believe the Sony based Nikons...I think the D7000 was the first...have linear noise in the sensor, such that at least to an extent, you can up ISO or push the image and you get the same noise levels...or something like that. I couldn't find a link.
I underexposed this shot by accident while we were testing lights.
ISO 100, as exposed, simply converted in LR:
Here it is pushed by FIVE stops:
Would I have gotten a better shot at ISO 3200? I don't think so. Well, maybe just a bit.
(Of course, this was simply a mistake.)
Message edited by author 2014-07-07 19:06:40. |
|
|
|
07/07/2014 03:14:37 PM · #6 |
Originally posted by LanndonKane:
that's a great answer. i forgot about the disadvantages of underexposure in terms of tone levels.
but i did hear that it was a bit of a myth that you lose half your tone values by underexposing in this way (or at least when worded in this way; maybe "bit depth" is more accurate). Could you clarify further about exactly what this means? |
To expand a little, a 1-stop exposure reduction relates to 1 bit. With each 1-bit reduction in data depth, half the values are lost. Example:
16-bit data depth = 2^16 values or 65536 values (per channel)
15-bit data depth = 2^15 or 32768 values
14-bit data depth = 2^14 or 16384
...etcetera.
Now, it is true that pretty much all 35mm and APS-C cameras don't use a full 16-bit data depth even if they store values as such. So, you may not really lose much the first stop you push. I believe you would lose something perceptible in the deep shadows, though. For High-ISO shooting, we probably don't really care so much about this. |
|
|
|
07/07/2014 03:22:02 PM · #7 |
As kirbic said, it's always better to get correct exposure in camera. Often the results may be similar, but if there is a noticeable difference, it will always favour the correctly exposed image.
You should also only use full iso stops for similar reasons (1/2 and 1/3 stops are just push/pulled in camera).
Message edited by author 2014-07-07 19:23:08. |
|
|
|
07/07/2014 04:45:40 PM · #8 |
Here's an interesting look at ISO versus noise.
Two shots...taken in a row in manual, with auto ISO bracketing. So all settings the same, but ISO cranked for the higher sensitivity shot.
If you bring down the exposure in LR to be the same, there's no difference.
This is a screen capture at 200% to emphasize the noise patterns. One is at ISO 2200...the other at ISO 8063.
 |
|
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: 04/05/2025 06:44:02 AM EDT.
