and help to keep the site free for all
just wondered ..
- Thread starter donut
- Start date
- Messages
- 2,666
- Name
- Alasdair Fowler
- Edit My Images
- Yes
Nope. The downside to long exposure using digitial is sensor noise.
mrgubby
Suspended / Banned
- Messages
- 7,988
- Name
- Brian
- Edit My Images
- Yes
It depends on whether sensor noise has a linear relationship with incident light.I suppose the heating effect on the sensor during long exposures would also affect the results.
If the resultis a non-linear relationship then it could be regarded as analagous to reciprosity failure .
Woodsy
Suspended / Banned
- Messages
- 7,676
- Name
- Jonathan
- Edit My Images
- No
I doubt the temp of the sensor increases during exposure (atleast no where near enough to make the difference in KbT large enough to alter the energy of each well), as the energy from each photon is not dissapated, it is used to increase the state of the quantum well in which the energy is stored, and is not passed into the surrounding medium. I believe the noise on CCD / CMOS is the result of the sensor simply not being at absolute zero for long exposure. For short exposure, noise is caused not because the sensor is more sensitive, but because the information from each pixel "well" is amplified to give the effect of increased sensitivity. However, the thermal (and other forms) noise is also amplified by equal amounts.
I believe this to be the case, I will be happy to stand corrected if not
. Of course, this does not answer the question about whether the relation is linear or not, but then again, is the noise dependent on the incident light? :shrug:
Reciprocity of film is due to the requirement of a certain amount of photons per unit time in order to change the state of each grain. If the intensity is not great enough, the partial change in each grain is not long lasting enough to hold the change in state. Digital sensors don't work this way. Though that being said I am not sure if they are 100% efficient in the way they work...
I believe this to be the case, I will be happy to stand corrected if not
. Of course, this does not answer the question about whether the relation is linear or not, but then again, is the noise dependent on the incident light? :shrug:Reciprocity of film is due to the requirement of a certain amount of photons per unit time in order to change the state of each grain. If the intensity is not great enough, the partial change in each grain is not long lasting enough to hold the change in state. Digital sensors don't work this way. Though that being said I am not sure if they are 100% efficient in the way they work...
OP
donut
Suspended / Banned
- Messages
- 3,088
- Edit My Images
- No
thanks all . as mk says the downside is more noise on digital ,as opposed to weird colours on film . no one medium can be all things , which is what shoots the digital v film debate down ( quickly steps in to say that is not what the op was about ) anyway woodsy we both got d50's . it was just an idle thought as i was pointing a sensor up into the sky last night hoping for a few geminids to go past ..
