Definition of ETTR

Started 6 months ago | Discussions thread
Flat view
Tom Axford Veteran Member • Posts: 5,499
Definition of ETTR

In a thread in the Beginners Questions, knickerhawk has questioned the commonly accepted definition of ETTR. I am not prepared to argue this question in a beginner's thread, so I have continued the discussion here.

knickerhawk wrote:

Tom Axford wrote:

Rather than argue with you, I'll refer you to the excellent article on ETTR on this website. It goes into considerable detail on how to do ETTR.

Gollywop, the author of the article, and I go way back and I'm a fan of his, and he's a fan of mine with respect to our mutual understanding of ETTR. See for instance his response to one of my posts here in a semi-technical thread about ETTR, where he wrote: Those wishing to understand the practical aspects of achieving ETTR would do very well to read [Knickerhawk's post], and read it carefully. This applies to newbies and apparently some oldies alike.Perhaps you should take Gollywop's advice as well.

Later in that same thread, I address the underlying ambiguity of the term "ETTR" and its evolving usage over times (and the associated implications regarding "ETTL"). You can read my post about it here, but in short I argued that the meaning of "ETTR" has morphed from its original application to relatively low DR conditions that support positive exposure compensation relative to normal JPEG metering to its use today to serve as a generalized synonym for optimized raw exposure. It's original proponents did NOT envision using the term as it is currently invoked. I've quoted two of the original proponents of the term (Jeff Schewe and Bruce Fraser) from their Real World Sharpening book to support my historical observation, wherein they caution against using "ETTR" in high DR situations. The only way it makes sense for world-class experts like Schewe and Fraser to caution against ETTR in such situations is because they were using the term to mean literally what it implies - that is, expose so as to move the histogram to the right relative to whatever default metering suggests. Any other strategy that results in the opposite histogram movement from the default metering simply and literally isn't "ETTR" for them even though it might be the optimal raw exposure strategy for the conditions. The meaning of the term has broadened over time.

The unfortunate result of the muddle that has come of the evolving usage of the specific term is that it leaves unsophisticated users (and even pretty knowledgeable photographers who think they have a solid grasp of the concepts) confused about its application and implications in high DR scenarios. This thread is proving to be yet another example of how easy it is to get tripped up by the unfortunate terminology and the ambiguous meaning of the acronyms of ETTR and ETTL. With that in mind, I interpreted (I believe correctly) that the original poster was asking about scenarios in which you might find it advisable to decrease exposure comp relative to normal JPEG-based metering of a scene, which is generally the opposite of what happens in "ETTR" (narrowly and literally defined as originally used and how unsophisticated users still think of it). I interpreted the OP to be talking literally and narrowly about situations that result in the histogram moving to the left relative to the normal JPEG-based metering instead of to the right. Sure, you can choose to define away the problem by pronouncing the OP's specific scenario as just another form of ETTR. But I prefer not to hide behind confusing terminology. Instead, I prefer to be more explicit about what's really going on, especially when trying to advise someone trying to understand the underlying issues.

In his article, gollywop never mentions ETTL, nor is it mentioned in the Wikipedia article on ETTR.

Wikipedia defines ETTR as:

"the technique of adjusting the exposure of an image as high as possible at base ISO (without causing unwanted saturation) to collect the maximum amount of light and thus get the optimum performance out of the digital image sensor"

That is the definition that I accept and that seems to me to be very widely accepted.

It is totally irrelevant what the camera's metered exposure is, and hence it is equally irrelevant whether the histogram moves to the right or to the left (from the camera's metered exposure) to achieve ETTR.

Flat view
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow