4
\$\begingroup\$

How does the exposure slider work in Adobe Camera Raw/Lightroom?

Naively I would expect the exposure slider to be a simple linear scaling (in the input color space, not in the working color space), but it's obviously not that. It's applying some sort of compression because you can increase it a lot without blowing up your highlights. Is this compression curve documented somewhere?

Related to that, do the numbers in the exposure slider have any absolute meaning? E.g. I have seen people claim -1.0 is one stop less exposure, but I don't think that's right. When I use graduated neutral density filters I use 0.6 and 0.9 grads (2 and 3 stops), but when I achieve the same effect in Adobe Camera Raw I rarely need to do more than -1.

Related questions:

What do all the settings do in Lightroom?

Does the Exposure Slider in Adobe Camera Raw Have Same Effect as in camera Exposure Compensation?

\$\endgroup\$
1

3 Answers 3

1
\$\begingroup\$

I don't have a proper answer yet, but I have a partial answer.

It seems the parameters in the DNG map directly to values in ACR, sort of making DNG an internal serialization format for ACR. DNG is documented, but there's no explanation how to compute these specific transformations (which also means any other DNG program will not render the same as ACR without reverse engineering ACR; quality archival format!).

Adobe also provides an SDK (unfortunately C++) for working with DNG images. It appears this SDK is pretty complete and will do DNG "exposure" adjustment. There is no reason why ACR would use the same library, or even the same algorithm, but it's the best we got so far.

I have not yet read the code in depth to answer my question precisely (the code is a mess), but I will do so and update this answer.

\$\endgroup\$
1
\$\begingroup\$

The Exposure slider in ACR does linear scaling. But if the image, post scaling, is adjusted non-linearly (S-curve, highlight compression, etc.), then additional non-linear effects accrue.

One way to demonstrate the meaning of the Exposure scale and that it is linear is to take two captures of the same scene with the second image underexposed one stop*. Then develop first image using arbitrary adjustments. Now use the same settings to develop the underexposed image but bump the Exposure slider by 1 stop. They will match. If the Exposure setting was not linear or the scale factor of 1 did not mean 1 stop the images would differ.

*Stops on cameras aren't often that accurate but are for longer exposure times when the aperture is constant. For instance, a 2 sec. exposure is exactly 1 stop above a 1 sec. exposure and most any camera will do so with an accuracy exceeding 99.9% while an exposure of 1/2000th v 1/4000th sec. can easily vary 10 or 20% because of manufacturing variations controlling the leaf separations slew rate.

\$\endgroup\$
0
\$\begingroup\$

Mine is also only (at best) a partial answer (and maybe not even a good one at that). Re: how you can move the slider quite a bit without blowing out the highlights - my understanding is that the exposure slider in Lightroom (and presumably in ACR as well) is controlling only the midtones, not the entire range of tones in the photo. At some point, moving up those midtones will force the highlights up enough to blow out, but until that point it seems to me there's quite a bit of latitude.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.