Quote Originally Posted by Medley
Really? So at what point did you add detail that wasn't captured by the camera's sensor? You didn't. You may have made the detail more readily noticeable, but you haven't ADDED anything.

If however, you reshoot the same scene at, say +2 ev (essentially exposing for shadows), then that area of the image will have more detail, due precisely to the sensor's non-linear nature. Yes, the lighter parts of the image may be blown out, but that doesn't matter, because you're going to substitute those shadows for the shadows in the original image, and discard the blown out areas. By doing so, you've increased the amount of detail in the resulting composite image. It now has more detail than the sensor was able to capture in the original image. You have ADDED detail to the original image.

But I'm not done. I now have to take my image that's exposed for shadow detail, and give those shadows the same approximate color and luminance of the shadows in the original image. In essence, I have to turn them back into shadows. THAT"S tone mapping, and it's an integral part of the HDR process.

But if you're not taking advantage of the non-linear nature of a digital sensor by shooting multiple exposures and blending them, then you're not adding any information, any detail, to the image. You're not doing HDR, you're just tone mapping.

TF, I understand and agree with what you're saying about the limitations of the sensor and output devices. And in that respect, no displayed or printed image is truly going to have a higher dynamic range. Our differences seem to come in how we overcome that problem. In tone mapping, you widen the dynamic range and redistribute the available information to make the image seem more realistic. In HDR, you add vast quantities of information that wasn't there before, than make strategic decisions on how and where to pare that information to fit within the dynamic range of the output device. The difference is that you can do so and still end up with an image that has more information than the camera's digital sensor is capable of recording in any single exposure.

As I said, the difference is subtle, but important.

- Joe U.
If it provides detail to areas that would have been blown out or black in the final image, I'm willing to let it be called HDR. I have seen others who insist single shot be called EDR - Extended Dynamic Range. ???

HDR IS a type of tone mapping. Now I understand that you are saying that all HDR (multi-shot) is doing is adding detail within the existing dynamic range. The primary purpose of HDR is to extend the dynamic range above and below the existing – information that would have previously been off the left or right of the histogram. It’s NOT due to “…the [linear] nature of a digital sensor…” (“non-linear” was a typo, correct?), it is providing information that was below the minimum sensitivity and/or above the maximum intensity range of the sensor. ‘Exposing to the right’ (see BlueRob’s post) provides more detail within the existing range, but that’s not HDR.

TF