Mad Hatter Imagery wrote:
What is the mechanical difference between taking a 1 minute long exposure and taking 60 one second exposures end to end and averaging them together in post?

In this experiment, are you planning on leaving all the other settings alone?

*edit:  And do you mean "adding" them together (not "averaging") ?

Mad Hatter Imagery wrote:
What is the mechanical difference between taking a 1 minute long exposure and taking 60 one second exposures end to end and averaging them together in post?

A one minute exposure is a one minute long exposure.

60 one second exposures averaged will result in a one second exposure.  Do you mean 60 exposures summed?

Mad Hatter Imagery wrote:
What is the mechanical difference between taking a 1 minute long exposure and taking 60 one second exposures end to end and averaging them together in post?

This a fairly common trick when shooting view cameras to shoot multiple exposures to get maximum DOF. But usually you are talking 2-4 exposures.

Mad Hatter Imagery wrote:
I mean one photo exposed for 60 seconds versus 60 photos exposed for one second each without delay between them and have all of them "averaged" together colorwise pixel by pixel. And yes lets assume the camera can be adjusted to produce the proper exposure and identical field of focus.

Would the resulting two photos (assuming they were taken at the same time and place in alternate universes that are also identical to one another) be the same?

under the same lighting conditions either using an nd filter for the 60 second exposure or higher iso for the 1 second exposures, you might lose some image quality with the nd vs extra noise in the high iso shots which may or may not be totally removed during the averaging, plus the differences i mentioned in my first reply.

Mad Hatter Imagery wrote:
I mean one photo exposed for 60 seconds versus 60 photos exposed for one second each without delay between them and have all of them "averaged" together colorwise pixel by pixel. And yes lets assume the camera can be adjusted to produce the proper exposure and identical field of focus.

Would the resulting two photos (assuming they were taken at the same time and place in alternate universes that are also identical to one another) be the same?

Mad Hatter Imagery wrote:
I mean one photo exposed for 60 seconds versus 60 photos exposed for one second each without delay between them and have all of them "averaged" together colorwise pixel by pixel. And yes lets assume the camera can be adjusted to produce the proper exposure and identical field of focus.

Would the resulting two photos (assuming they were taken at the same time and place in alternate universes that are also identical to one another) be the same?

Yes to answer your question.

The classic way of looking at is if you have an exposure with ISO 100 at f/16 at 1/60.

I want to shoot at f/22. I would have to do a double exposure at 1/60 at f/22 to equal 1/60 at f/16 (normal exposure 1/30 at f/22).

At /32 at a 1/60 you would need to expose the medium 4 times (normal exposure 1/15 at f/32).

At f/64 at 1/60 you would need to expose the medium 8 times (normal exposure 1/8 at f/64).

Any you should get the idea. Is it possible do 60 multiple exposures as opposed to one 60sec exposure yes is it practical not really.

The big thing is remember the relationship of f/stops shutter speeds and ISO. Pretty much the first thing covered in any basic photo class. If you do not understand this relationship I recommend learning it. It will expand your creative horizons and give much more creative freedom and control. It is the difference from playing guitar hero to actually making music with the guitar. http://www.cambridgeincolour.com/tutori … posure.htm

To give you an idea of this I used this concept for this lunar eclipse shot with a 4x5 viewcamera years ago. I did multiple exposures on the same film.


moon by FullMetalPhotographer, on Flickr

Mad Hatter Imagery wrote:
I mean one photo exposed for 60 seconds versus 60 photos exposed for one second each without delay between them and have all of them "averaged" together colorwise pixel by pixel. And yes lets assume the camera can be adjusted to produce the proper exposure and identical field of focus.

Would the resulting two photos (assuming they were taken at the same time and place in alternate universes that are also identical to one another) be the same?

No, they will not be the same.

Think of this way.

You have a bucket.  You fill it for 60 seconds.  You have plenty of water.

Now you fill it for only 1 second, toss the water out, start again. You do this 60 times.  When you're done, you'll have very little water in the bucket.

The only way your idea would work is if you expose the same frame 60 times and you'll need a camera that supports multiple exposures and to that quantity.

ETA: I reserve the right to be incorrect.

Christopher Hartman wrote:
No, they will not be the same.

Think of this way.

You have a bucket.  You fill it for 60 seconds.  You have plenty of water.

Now you fill it for only 1 second, toss the water out, start again. You do this 60 times.  When you're done, you'll have very little water in the bucket.

The only way your idea would work is if you expose the same frame 60 times and you'll need a camera that supports multiple exposures and to that quantity.

ETA: I reserve the right to be incorrect.

Don't toss the water out

Christopher Hartman wrote:
Again, I could be completely wrong.


But if you shoot the same thing 60 times and there are no changes in settings or lighting...then why not just take the photo one time and then make 60 copies?  wouldn't that accomplish the same thing?

Trying to wrap my brain around this...

I depends what you are photographing, photons flux is not always constant.

Take my photograph of the andromeda galaxy, if I give you  a single one of the 20 frames I used to build that image and then you make 20 copies  you wouldn't end up with an image the same as the one I posted.

You can try if you want, this is frame two out of 20 in that series.

Christopher Hartman wrote:
Again, I could be completely wrong.


But if you shoot the same thing 60 times and there are no changes in settings or lighting...then why not just take the photo one time and then make 60 copies?  wouldn't that accomplish the same thing?

Trying to wrap my brain around this...

Agreed.

OP,

If we were talking about exposing one sheet of film, 60 times for a second each on the same sheet of film, then yes, it would be equal to one 60 second exposure.  Depending on film, you might have to take reciprocity failure into account if going for longer than a minute.

But that is a relatively linear process.  I have no idea how you would "average" that in post using 60 different photos (which doesn't mean it can't be done, just that I don't understand how or why you would do it)? 

I've shot night scenes on a digital camera (old tech, not the new high iso wonders) at ISO 100 and had perfectly clean images at 30sec.  Same with my digital back, which gets really noisy after ISO 400.

So, putting the theoretical aside for a moment, what are you actually trying to accomplish?

HV images wrote:

I depends what you are photographing, photons flux is not always constant.

Take my photograph of the andromeda galaxy, if I give you  a single one of the 20 frames I used to build that image and then you make 20 copies  you wouldn't end up with an image the same as the one I posted.

You can try if you want, this is frame two out of 20 in that series.

Very interesting.  Can you explain the digital process in a bit more detail?

Mad Hatter Imagery wrote:
What is the mechanical difference between taking a 1 minute long exposure and taking 60 one second exposures end to end and averaging them together in post?

Y'know -- I'm not sure I understand the question.  Most importantly, are we talking about a still subject matter, or is the subject of the image moving?  If the camera is on a sturdy tripod and is locked down firmly and we are using a cable release, there should be very little difference between 60 1-second exposures & 1 60-second exposures -- in theory.

But...

...  If we are talking about a DSLR camera, taking multiple 1 second exposures means that there is more internal camera vibration, caused by the mirror flipping up & down 60 times.  A single 60 second exposure will have only one instance of a mirror movement.

...  In addition, cocking the shutter 60 times might introduce slight changes in the camera's position.

...  There are also other internal moving parts to a camera (e.g. stopping down the aperture) that can also introduce subtle movements & vibrations.

...  For the above three reasons, I would expect the single 60 second exposure to be sharper than the 60 1-second multiple exposures.

...  It should be mentioned that if we are talking about a minute exposure, we are talking about a low light situation.  Some digital sensors perform better (or at least differently) in low light when compared to others.

...  If we are talking about film (and maybe even about digital) -- that's a whole nudder animal.  In film, there is a condition called reciprocity failure -- essentially, during long exposures, some light sensitive materials (film & sensors) lose sensitivity & require compensation.  For example, if your light meter says that you can make an exposure at f/5.6 for one minute, you might have to use a much longer exposure to compensate.  Go & do a search for "reciprocity failure photography" for more information.

HV images wrote:

I depends what you are photographing, photons flux is not always constant.

Take my photograph of the andromeda galaxy, if I give you  a single one of the 20 frames I used to build that image and then you make 20 copies  you wouldn't end up with an image the same as the one I posted.

You can try if you want, this is frame two out of 20 in that series.

[img]http://www.hvalladares.com/mm/andromeda1.jpg[img]

It still hurts my brain, but I have heard of what you did.  I wasn't sure that is what the OP is talking about though.  I was reading his question as a more "general" all purpose question.

Other than astrophotography, does it work in other areas?

Giacomo Cirrincioni wrote:

Very interesting.  Can you explain the digital process in a bit more detail?

It is a bit lenghty but I can point you to a nice tutorial, it doesn't go into the why and how but it gives you a good insight into the digital process: http://digital-photography-school.com/h … otography/

I Hope it helps.

Giacomo Cirrincioni wrote:
Very interesting.  Can you explain the digital process in a bit more detail?

HV images wrote:
It is a bit lenghty but I can point you to a nice tutorial, it doesn't go into the why and how but it gives you a good insight into the digital process: http://digital-photography-school.com/h … otography/

I Hope it helps.

Very cool, thank you!

Christopher Hartman wrote:
Again, I could be completely wrong.


But if you shoot the same thing 60 times and there are no changes in settings or lighting...then why not just take the photo one time and then make 60 copies?  wouldn't that accomplish the same thing?

Trying to wrap my brain around this...

No.
Because if you duplicated the image, the noise would be the same in all your copies and would add. If you have 60 individual images, the noise, being random, will be different in each one and  will largely cancel itself out when you average them.

Christopher Hartman wrote:
Other than astrophotography, does it work in other areas?

There was a shot in ID4 with the Empire State Building about to explode that was excessively grainy, but the camera wasn't moving, so I had the compositor make each frame an average of the preceding 30 frames and that removed the grain because noise is random deviations both above and below the average ideal image.

Then when the building exploded we cut to the original grainy footage but it didn't matter because shrapnel was flying everywhere anyway.

Mad Hatter Imagery wrote:
What is the mechanical difference between taking a 1 minute long exposure and taking 60 one second exposures end to end and averaging them together in post?

Sometimes you just don't have one long exposure. Making sense of a license plate on a parked car in crappy surveillance video footage would be a non-creative example.

I haven't actually compared the noise in a long exposure at low ISO with the noise in many high ISO photos.

Christopher Hartman wrote:
ETA: I reserve the right to be incorrect.

Can I borrow/steal this?

Looknsee Photography wrote:
Y'know -- I'm not sure I understand the question.  Most importantly, are we talking about a still subject matter, or is the subject of the image moving?  If the camera is on a sturdy tripod and is locked down firmly and we are using a cable release, there should be very little difference between 60 1-second exposures & 1 60-second exposures -- in theory.

But...

...  If we are talking about a DSLR camera, taking multiple 1 second exposures means that there is more internal camera vibration, caused by the mirror flipping up & down 60 times.  A single 60 second exposure will have only one instance of a mirror movement.

...  In addition, cocking the shutter 60 times might introduce slight changes in the camera's position.

...  There are also other internal moving parts to a camera (e.g. stopping down the aperture) that can also introduce subtle movements & vibrations.

...  For the above three reasons, I would expect the single 60 second exposure to be sharper than the 60 1-second multiple exposures.

...  It should be mentioned that if we are talking about a minute exposure, we are talking about a low light situation.  Some digital sensors perform better (or at least differently) in low light when compared to others.

These are probably the only differences that I can see. Ignoring the fact that the OP used "averaging", which would create an apples and oranges situation, perhaps cumulative is the proper term or additive.

Well done.

Herman Surkis wrote:
These are probably the only differences that I can see. Ignoring the fact that the OP used "averaging", which would create an apples and oranges situation, perhaps cumulative is the proper term or additive.

Well done.

Averaging is just accumulating with an exposure adjustment at the end.

You add N things together, and then make the sum 1/N as bright.

NothingIsRealButTheGirl wrote:
http://www.digitalartform.com/archives/ … oise_1.jpg

Here's one image on a D200 at ISO 1600 which is noisy on that camera.

http://www.digitalartform.com/archives/ … oise_8.jpg

Here are 8 such images averaged.

http://www.digitalartform.com/archives/ … noise.html

I was thinking of trying this myself when I got off work today, to see what would happen for both averaging and adding images together.  I would have written a python script to do it though because GUI photo editors are for losers.

Schlake wrote:
I was thinking of trying this myself when I got off work today, to see what would happen for both averaging and adding images together.  I would have written a python script to do it though because GUI photo editors are for losers.

One old school (pre 'image stack') way to do this is:

Bottom layer at 1/1 = 100% opacity
The layer above that 1/2 = 50% opacity
The one above that at 1/3 = 33% opacity
The one above that at 1/4 = 25% opacity
The one above that at 1/5 = 20% opacity
.
.
.
etc

The math works out to normal averaging.

Schlake wrote:
I would have written a python script to do it though because GUI photo editors are for losers.

That's why I use the GUI.

btw, if you go into 32-bits you can add using linear dodge blend mode way above 100% and then use the exposure adjustment layer to do the final 'divide'

sort of like this:
http://www.youtube.com/watch?v=Ag4e32VfDvE

fullmetalphotographer wrote:
There are cameras like the nikon D3 that you can do multiple exposures with.

my k-30 does that. up to 9 frames, either additive or averaged.