How much shutter-timing variation causes a visible exposure difference?

Asked 5/21/2013

3 views

2 answers

0

If shutter speed varies slightly from shot to shot while aperture and ISO stay fixed, how much variation is likely to produce a noticeable brightness change between otherwise identical photos? For example, if a nominal 1/60 s exposure might actually be about 1/59 to 1/61, is there a practical threshold in percent or EV where viewers will start to notice the difference, especially when comparing images side by side or in sequence?

Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0

Photography Stack Exchange contributor

13y ago

2 Answers

3

Because the exposure factors of ISO, shutter speed, and aperture are interchangeable in terms of final exposure, you can experiment with this yourself by loading a sample image into a photo processing application like Darktable and playing with the Exposure slider.

The bad news is that the human eye is really good at this, at least when the images are shown in sequence. I can easily tell a difference of about 0.05 EV, and even smaller changes are apparent if I'm paying attention.

On the other hand, were the photos to be displayed across the room from each other, I bet most non-photographers wouldn't notice a difference of up to half EV.

(Note that 0.05EV is not a percentage, since the EV scale is logarithmic; it's something like 3.5%. And half an EV is 41%. I think you might be able to get away with 1%, but it'll depend on a number of factors; it's probably best to decide for yourself what's acceptable.)

Originally by user1943. Source · Licensed CC BY-SA 4.0

user1943

13y ago

0

AI Answer

Generated from our catalog & community — verify before relying on it.

There isn’t a single hard cutoff, because visibility depends on how the images are viewed. The useful way to think about it is in EV, not percent, since exposure is logarithmic.

From the answers: side-by-side or sequential images make small brightness differences much easier to see. Changes around 0.05 EV can be noticeable to a careful viewer, while people are much less sensitive when images are separated; in that case, even much larger differences may go unnoticed.

Your example of 1/59 to 1/61 around a nominal 1/60 is only about ±1.7% in time, which is roughly ±0.025 EV. That’s very small. The total spread between the brightest and darkest shots there is about 0.05 EV, which may be detectable in direct comparison, but is subtle.

So as a rule of thumb:

  • for matched images shown side by side or as a sequence, keep variation under about 0.05 EV if possible
  • if images are viewed separately, much larger differences may be acceptable

If timing jitter stays around 1–2%, it will usually be minor; around 5–10% it becomes increasingly likely to be visible in direct comparisons.

UniqueBot

AI

13y ago

Your Answer