Theory of Exposure

ernesto

ernesto

Well-known
Local time
2:02 PM
Joined
Nov 5, 2007
Messages
254
Hello!

I have to take a picture of an architectural space, using multiple exposures of 1/30 second each. They will be 10 of them. My problem is calculating the resulting total exposure time. From a mathematical point of view, 10 x 1/30 = 1/3 second. But I have read that Film does not work that way due to ‘reciprocity failure’, which is a phenomenon experienced when shooting extra long exposures with film. The reason for such reciprocity failure with film is that the longer the exposure, the less effective the film is in recording the light and so the exposure length needs to be increased, but I have no idea how much.
https://en.wikipedia.org/wiki/Reciprocity_(photography)
I wonder if someone could help me with this problem.
Thanks in advance!

Ernesto
 
This will depend on your individual film material and the developer combination. You wil have to do some test shots to find the best result for your purpose.
But what is the reason for combining 10 1/30 second multi exposures instead of using a single 1/3 exposure ... capturing people moving in the scene?
 
A quick search on fuji film (for instance) will show the wild variation between Velvia and Provia . A mathematical holy grail to cover all films is a pipe dream . Peter
 
Test with a B&W film like Fujifilm Neopan 100 to see if you like the effect and then adjust for the specific color film???

Fujifilm Neopan Acros 100 Reciprocity Characteristics: No exposure compensation is required for exposures at shutter speeds of less than 120 seconds.
 
In theory, you are right, but there will be no reciprocity if both the partial times and the sum are within the safe limits of the reciprocity thresholds. The figures you are operating with are all well within the reciprocity-free zone for every currently available film. YMMV depending on the film, but even the most vulnerable films available today are rated for no reciprocity between 1-1/1000s, and the majority will even handle 10-1/10,000s without correction factors.

Something like ten exposures of a second each would be a different matter, with many films at least. And so would be a hundred of 1/30,000.

Beware that multi exposures may amplify (usually invisibly small) flaws with the shutter that are constant but happen at each exposure (like shutter bounce or constant time deviations).
 
If your final exposure is going to be 1/3 of a sec you shouldn't have to worry about reciprocity failure. For me I wouldn't even worry about it until you need to make an exposure of 3 sec or more.

Lastly, long exposures on film don't need to be nearly as accurate as you think. For example When shooting bulb I'll never get it "exactly" 10 seconds, but will be a third or half second late. It doesn't matter. When shooting long exposures you have wiggle room and the beauty of film is you still have dynamic range on your side.
 
Ernesto, I think the other answers here are right on the mark. In the range you are talking about, exposure should accumulate linearly. No reciprocity problem in this range.
 
Since all of the exposures are short there is no reciprocity failure. You should use the calculated total time without applying and correction for reciprocity.
 
The comments on reciprocity are correct but can I just clarify that your meter shows a total exposure of 1/3 second is required which you are going to achieve by 10 exposures of 1/30 sec?
This may seem obvious and for that I apologise but if the scene requires 1/30 sec adding more exposures will cause overexposure quite quickly, I may well be wrong but the way you have laid out your calculation worries me on this point.

Away from theory you will find that progressive exposures will tend to burn out highlights more than a single exposure when calculated this way. It is common practice in those using this technique to slightly underexpose the first exposure and progressively increase underexposure with each additional shot.

By giving an exposure to the silver halide grains you are increasing its sensitivity to further exposure. An area that is just registering to light on film will have halide not exposed sufficiently to "trigger" a response in those areas a subsequent exposure that again is below the normal threshold will however find the threshold lowered by the previous exposure and will "trigger". This can give an overall exposure greater than the the one calculated by the simple mathematics of addition hence the progressive underexposure recommended. It is difficult to be accurate as the time interval between exposures has a large effect, pausing allows the threshold to reset in some emulsions, but without extensive work is hard to predict.

Don't fret about this just Bracket, but I do wonder what the requirements of an architectural space for multiple exposures are unless you are moving lights around, "painting" with lights can be a very useful technique but is usually done on a single long exposure because of the potential shutter effects noted above or by using flash. Particularly useful in caving work, a friend of mind did extensively, never persuaded me down there though.

In reciprocity failure the threshold effect is again the cause, light photons hitting the halide are not sufficient to trigger a response immediately and the re-set occurs quickly so giving more exposure only allows more photons to hit the same site without triggering, the variable reset is why emulsions vary in reciprocity. An image is eventually formed due to the random nature of the energy distribution of the photons, eventually sufficient photons of high energy will "trigger" the response.
The incorporation of sensitisers of higher efficiency and tighter control over crystal shape has reduced reciprocity effects in many modern films.
 
icebear said:
This will depend on your individual film material and the developer combination. You wil have to do some test shots to find the best result for your purpose.
But what is the reason for combining 10 1/30 second multi exposures instead of using a single 1/3 exposure ... capturing people moving in the scene?
Click to expand...

I am using a Noblex 135s panoramic camera (for the first time) which slowest speed is 1/30 second, and I am planning to photograph a dark church space.
 
BLKRCAT said:
If your final exposure is going to be 1/3 of a sec you shouldn't have to worry about reciprocity failure. For me I wouldn't even worry about it until you need to make an exposure of 3 sec or more.

Lastly, long exposures on film don't need to be nearly as accurate as you think. For example When shooting bulb I'll never get it "exactly" 10 seconds, but will be a third or half second late. It doesn't matter. When shooting long exposures you have wiggle room and the beauty of film is you still have dynamic range on your side.
Click to expand...

I see... Thannks!
I guess I should try how it looks without any compensation factor.

E
 
ChrisLivsey said:
The comments on reciprocity are correct but can I just clarify that your meter shows a total exposure of 1/3 second is required which you are going to achieve by 10 exposures of 1/30 sec?
This may seem obvious and for that I apologise but if the scene requires 1/30 sec adding more exposures will cause overexposure quite quickly, I may well be wrong but the way you have laid out your calculation worries me on this point.

Away from theory you will find that progressive exposures will tend to burn out highlights more than a single exposure when calculated this way. It is common practice in those using this technique to slightly underexpose the first exposure and progressively increase underexposure with each additional shot.

By giving an exposure to the silver halide grains you are increasing its sensitivity to further exposure. An area that is just registering to light on film will have halide not exposed sufficiently to "trigger" a response in those areas a subsequent exposure that again is below the normal threshold will however find the threshold lowered by the previous exposure and will "trigger". This can give an overall exposure greater than the the one calculated by the simple mathematics of addition hence the progressive underexposure recommended. It is difficult to be accurate as the time interval between exposures has a large effect, pausing allows the threshold to reset in some emulsions, but without extensive work is hard to predict.

Don't fret about this just Bracket, but I do wonder what the requirements of an architectural space for multiple exposures are unless you are moving lights around, "painting" with lights can be a very useful technique but is usually done on a single long exposure because of the potential shutter effects noted above or by using flash. Particularly useful in caving work, a friend of mind did extensively, never persuaded me down there though.

In reciprocity failure the threshold effect is again the cause, light photons hitting the halide are not sufficient to trigger a response immediately and the re-set occurs quickly so giving more exposure only allows more photons to hit the same site without triggering, the variable reset is why emulsions vary in reciprocity. An image is eventually formed due to the random nature of the energy distribution of the photons, eventually sufficient photons of high energy will "trigger" the response.
The incorporation of sensitisers of higher efficiency and tighter control over crystal shape has reduced reciprocity effects in many modern films.
Click to expand...

Interesting ChrisLivsey
My problem is that I am using a swing lens camera, and I cannot use anything close to Bulb nor even 1/3 of a second exposure.
It makes sense that the first exposure would affect film in a more intense way compared to the tenth exposure, which is hitting an already exposed material...
Anyway I expect that in my case, this could not be a problem since the scene will remain unchanged.
We will see after development...
 
Thanks you all.
I love the idea that reciprocity failure should not bother in my case, just for the reason it will be easier to handle in the field...
Anyway I will post the results soon.
Thanks again!

E
 
Yes please post I would love to see the results.
As it takes around 3 seconds, I think, for the 1/30th slit to cover the film you are going to need that static subject and 2 frames a minute if bracketing, good luck and enjoy.
 
Dwig said:
Since all of the exposures are short there is no reciprocity failure. You should use the calculated total time without applying and correction for reciprocity.
Click to expand...

Short individual exposures are definitely no guarantee for no reciprocity failure.

Reciprocity failure is really about registering low light levels on film and not about exposure time by itself. If you had a multiple exposure of 100 exposures, each with 1/100s (1s of total exposure), the reciprocity failure will definitely not be less than a single 1s exposure. It will probably be more.

So, if the result has to be perfect on the day, a prior test is required.

Or just use Fuji Provia 100F or Acros 100 :)
 
You must log in or register to reply here.
Back
Top Bottom