ISO Invariance and exposure decisions

Started Aug 10, 2017 | Discussions thread
ForumParentFirstPrevious
Flat view
Bill Janes Senior Member • Posts: 1,807
ISO Invariance and exposure decisions
1

There has been much discussion on the net on how one can in limited light situations protect the highlights by shooting dark and boosting brightness in the raw converter during post. This strategy works best with an ISO-invariant camera.

An iso-invariant camera is one where read noise remains constant over the range of ISO values, and most recent Sony sensors are in this category. Partially ISO-invariant sensors become so above a certain ISO, often around 1600, and include many Canon cameras. A brief discussion of the topic can be found here.

Emil Martinec gives a comprehensive explanation of this topic in his excellent article on sensor performance (look at S/N and Exposure Decisions just beyond Fig 20 in the treatise). A somewhat easier to understand article for the less technically inclined may be found here .

Charts on Bill Claff's site are helpful in determining if and when one's camera is ISO-invariant. In this chart, the sensor is ISO-invariant when the read noise in ADUs vs ISO becomes linear.

Now assume we are in a low light situation where we need an aperture narrow enough for adequate depth of field and a shutter speed high enough to freeze motion and control camera shake. We are navigating between Scylla and Charybdis in this situation and can not achieve the ideal. but how does one reach a suitable compromise?

One method is to determine the highest ISO one's camera that gives an acceptable signal to noise ratio with a normal exposure. This is subjective, but I will assume ISO 3200 for my Nikon D800e. In manual exposure mode select what you think is a suitable compromise for an ETTR exposure at ISO 3200. This exposure determines how many photons reach the sensor (exposure is measured in lux seconds and is determined by the aperture and shutter speed, and has nothing to do with the ISO plugged into the metering system to derive those values). At this point, some photographers would use -0.5 or -1 EV exposure compensation to protect the highlights in difficult lighting situations and then compensate in post, but this comes at the cost of decreased SNR in an already marginal situation.

A better solution in shooting dark and compensating in post would be to shoot at the metered shutter speed and aperture, but to decrease the ISO setting by one or 2 stops. This decreased ISO will reduce ISO amplification and protect the highlights with no loss of SNR. With a truly ISO-invariant camera, one could shoot at base ISO, but this would result in a very dark image on the LCD preview and render chimping difficult or useless. Also, many raw converts do not handle extreme brightening in post with introducing hue shifts or changes in contrast.

I invite discussion by other photographers how they handle this situation.

Regards,

Bill Janes

Nikon D800E
If you believe there are incorrect tags, please send us this post using our feedback form.
ForumParentFirstPrevious
Flat view
Post (hide subjects) Posted by
(unknown member)
(unknown member)
(unknown member)
(unknown member)
ForumParentFirstPrevious
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow