moreammo
Established
I took this shot several times and in order to properly expose it i made it too noisy due to the long exposure. The longer exposure time also changed the dynamic of the photo through cloud and star movement. I tried increasing ISO but that made the noise issue worse... i decided to shoot at ISO 160 and underexpose then try and pull back the detail. By underexposing i was able to keep the clouds and stars as i like them and to my surprise at iso 160 even after pulling it back almost two and half stops i gained no distracting noise. further more this was intended to be black and white but the camp fires lit the tree in such a way that i enjoy it in color.
garner tree by jon.manes, on Flickr
M9 35 Summilux F8 / 32 seconds / ISO 160
an interesting find to me thought i would share.
-Jon
garner tree by jon.manes, on Flickr
M9 35 Summilux F8 / 32 seconds / ISO 160
an interesting find to me thought i would share.
-Jon
DamenS
Well-known
Good one. You can actually take even shorter exposures and then "stack them" in photo editing software such as Photoshop - that means you wouldn't need to pull the exposure back by 2 and a half stops, and any corresponding "noise" will be substantially lessened.
moreammo
Established
Thanks for the tip! i'll have to try that
willie_901
Veteran
Under exposure is rarely benefits digital imaging.
Under exposure can not reduce noise. The noise can be considered constant for a given ISO (up to the electronic maximum). But the the number of photons recorded (the signal... or the information we want) depends only on the exposure. Intentionally reducing the signal by under exposure must reduce the signal to noise ratio compared to the optimal exposure. Of course this assumes the sensor does not overheat which would increase the overall noise at slow shutter speeds at any ISO setting. Overheating is minimized by increasing the ISO to shorten the shutter speed. Many contemporary cameras employ heat sinks and other heat dissipation techniques so video recording is practical. Still photographers benefit because long shutter speeds do not dramatically increase thermal noise.
Optimal exposure is when the the sensor sites count as many photons as possible (using the balls in a bucket analogy) except for highlight details unimportant to the scene of interest. This gives the maximum signal to noise ratio possible for all the sensor sites. While the dynamic range decreases as electronic ISO increases, the signal to noise ratio can remain constant as long as the exposure is optimal. I suspect increasing the ISO caused problems for you because of a decrease in dynamic range. I think you could have increased exposure to the point where brightest part of the clouds were ever so slightly overexposed. This would maximize the signal in the shadow regions.
Theory and empirical examples that describe this in detail are found here:
http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/
Under exposure can not reduce noise. The noise can be considered constant for a given ISO (up to the electronic maximum). But the the number of photons recorded (the signal... or the information we want) depends only on the exposure. Intentionally reducing the signal by under exposure must reduce the signal to noise ratio compared to the optimal exposure. Of course this assumes the sensor does not overheat which would increase the overall noise at slow shutter speeds at any ISO setting. Overheating is minimized by increasing the ISO to shorten the shutter speed. Many contemporary cameras employ heat sinks and other heat dissipation techniques so video recording is practical. Still photographers benefit because long shutter speeds do not dramatically increase thermal noise.
Optimal exposure is when the the sensor sites count as many photons as possible (using the balls in a bucket analogy) except for highlight details unimportant to the scene of interest. This gives the maximum signal to noise ratio possible for all the sensor sites. While the dynamic range decreases as electronic ISO increases, the signal to noise ratio can remain constant as long as the exposure is optimal. I suspect increasing the ISO caused problems for you because of a decrease in dynamic range. I think you could have increased exposure to the point where brightest part of the clouds were ever so slightly overexposed. This would maximize the signal in the shadow regions.
Theory and empirical examples that describe this in detail are found here:
http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/