1-Image HDR

[Helen B]

I'd be very interested in seeing a single exposure with a 15-stop exposure range, provided you have one to show. Details of the shot, including meter readings and development scheme would be greatly appreciated.

I have none of my notes or examples for B&W here in NYC, but as part of the D3 vs film evaluation I will re-do some of the tests. I used to use Xtol 1+5 and Rodinal 1+300 for contraction. A range of 15 stops is well beyond what one would normally require.

Here are the results of my test of the previous version of Portra 160NC. I didn't give it enough exposure to reach the shoulder, but you can see that it has at least 13 stops of range above the toe. That is more than enough for most scenes - ie the film can cope with a greater brightness range than most scenes, and using multiple exposures will not increase the recorded dynamic range.

Best,
Helen

 

[Garbz]

and using multiple exposures will not increase the recorded dynamic range.

True for any single image but not a combined version of the three. Suppose you have a scene with 10 stops of dynamic range you want to extract detail from. And for argument's sake just suppose our digital sensors range in the number i said earlier of 6 stops.

Exp 1: EV range of 3-8
Exp 2: EV range of 1-6
Exp 3: EV range of 5-10

Now combine all of these into a 32bit hdr image and you won't be able to display let alone print the image without some kind of tonal compression. Apply tonemapping and volah, you've now extracted detail from the 10th and the 1st exposure value as it was all recorded in separate images. That is extra dynamic range in the purest sense of the word.

 

[Helen B]

Garbz,

I'm not in any disagreement with your general approach, but we are looking at it in two different ways.

Helen: "and using multiple exposures will not increase the recorded dynamic range."

Garbz: "True for any single image but not a combined version of the three."

I was referring to the case where one exposure captures the entire dynamic range on the straight-line portion of the response curve. The full sentence you quote from was: "That is more than enough for most scenes - ie the film can cope with a greater brightness range than most scenes, and using multiple exposures will not increase the recorded dynamic range."

Remember, Max's proposition is that a single image can never be 'true HDR' - I'm just explaining (and demonstrating) why there are cases when a single exposure can be true HDR. I'm not disputing the idea that you need multiple exposures when the scene brightness range exceeds the capabilities of the film or sensor.

Best,
Helen

 

[Garbz]

Ahhh gotchya. I guess it depends on the definition of HDR. Many of the digital guys on this forum would sware it means photographing more dynamic range than the exposure latitude of the digital sensor. But we're on the same page now. :thumbup: