Iso Idea/Question.

[ravncat]

Heylo,
Not sure where the right spot for this is - and, I haven`t really had the chance to go out and test the idea ..

In general, as we increase our ISO we decrease the dynamic range captured, as well as the "color depth" and we increase noise. For the most part, conventional digital photography seems to assume low ISO whenever you can - high ISO when you don`t have the light. (although there are certainly those who are using Iso to create noise for creative purposes, and there are also a whole host of little "niggles" to deal with when it comes to ISO discussions, I`d iike to leave those out for now. )

So - Is there benefit or reason to use the ISO in concert with reproducing the image across a tonal range? Say for example - That we have two scenes - one is high contrast - and covers 10 or 11 stops of range from the shadows to the highlights (that are important to us) and one scene that is lower contrast and covers say a 4 or 5 stop range from shadow to hilights. Both scenes are well lit.

Normally we would use a low ISO and shoot both scenes - and some would maybe manually adjust exposure for the second scene in accordance with ETTR (exposure to the right) practice; to be reprocessed in post. Would there be benefit to instead capture the second scene with a Higher ISO? The Reasoning being that, at say, some higher ISO we capture a smaller dynamic range, and we can fit our lower contrast capture through the entirety of that smaller dynamic range?

I need to do some shooting on this - I just haven`t had the time lately. I`m curious about the results, as well as what others think. I don`t really have any conclusions yet to share myself.

[wulf]

An interesting theory but I'm not sure it would get much chance to breathe in practice. The side effects of the other two corners of the exposure triangle (depth of field with aperture and blurring / freezing motion with shutter speed) have a more obvious effect on the final result. The only side effect I have noticed with high ISO is that it creates some grain and some posterisation, normally ugly, particularly in areas that are still under-exposed.

Therefore, I tend to think of aperture and shutter speed as the two main variables although I am less afraid of high ISO than I used to be. You might be able to exploit your theory if you were in a situation with good control of the lighting but I find that most of my shooting situtations are too constrained to give it much thought.

Wulf

[OsmosisStudios]

The more information the better: more dynamic range will always be better than less, even if you dont need it. It would also mean less noise, which is always a plus.

[autofocus]

Just a little comment about dynamic range of 10 - 11 stops. We humans can discern all the subtleties in that range, but the camera cannot. At best, most cameras today can only see about 5 -7 stops of dynamic range (unless you process HDR images)

[inkista]

I'm not sure I buy the idea that higher iso necessarily limits your dynamic range. Changes it, yes, but I don't think it's the kind of linear progression you're positing here. As evidence, here's the dpreview testing for DR on the nikon D300 by iso setting:



Note that it's not a simple matter of a lower iso having a more extended range. While iso 100 has more range at the lower end, it's got less at the higher end. And, of course, whether the artificial processed iso levels (is the base iso of the D300 iso 200? or iso 100?), like iso 50 on my 5D2, brings in a whole 'nother can of worms.

And hell, you could probably increase dynamic range just by lowering contrast.

As Wulf says, I doubt it matters much in practice.