[jeffshaddix]

Unwrapping ADC rollover isn't a novel idea, but perhaps their processing algorithm is. The problem is that no matter what your algorithm, you cannot guarantee proper recovery in the presence of strong noise or strong contrast edges (with multiple rollovers). And in those cases, the artifacts are unbearable.

It would be a nice option to let raw files store rollover counts instead of saturating (maybe they already do?), and have a Photoshop plugin to try to unwrap rollover, but this approach is not reliable enough to go further in my opinion.

[Simen1]

How limitless is the non overexposure feature? Can i take a landscape image with the sun in the frame, and push the exposure to see sun spots and filaments? Or maybe more likely a nightscape with both the milky way correctly exposed and the moon pushed to see features on it in a single exposure?

How many bits of color depth are captured and stored?

[Rondec]

How limitless is the non overexposure feature? Can i take a landscape image with the sun in the frame, and push the exposure to see sun spots and filaments? Or maybe more likely a nightscape with both the milky way correctly exposed and the moon pushed to see features on it in a single exposure?

How many bits of color depth are captured and stored?

The problem you run into when the sun is in the frame is flare and this will do nothing for that -- just a basic loss of contrast around super-bright objects in your frame. It looks to me, that it would be most useful for high dynamic range situations, where there is plenty of light. Currently I tend to underexpose to protect the highlights, but this sort of camera would make that less necessary, as it would take a lot to blow out the highlights.

[pathdoc]

Vs

Interesting. The first one has almost a drawn or painted appearance. It suits that particular image nicely IMO, but might not do so well for others.

[vonBaloney]

The problem you run into when the sun is in the frame is flare and this will do nothing for that -- just a basic loss of contrast around super-bright objects in your frame. It looks to me, that it would be most useful for high dynamic range situations, where there is plenty of light. Currently I tend to underexpose to protect the highlights, but this sort of camera would make that less necessary, as it would take a lot to blow out the highlights.

Yeah, in most cases I don't think it will be that big of a deal in terms of making "never before possible images" or something. Like I said before, it would be like a return to negative film -- right now we shoot on "direct positive", like slides. Of course we have much more leeway than with slides, but still blown highlights can never be recovered. Whereas with negative film, you didn't have to worry about the highlights so much (not unlimited, but 2-4 stops leeway) and could expose for the shadows. So this will be a welcome tech if they get it right (and there will be others ways to achieve it I'm sure). But it isn't revolutionary for standard use because we had it already for 100 years and we'd just be getting it back but with even more latitude. I'm sure you could also push the envelope and do some weird things with it, but that would involve long exposures and what not.

So for normal photography, this tech is not about avoiding overexposure as much as it is about avoiding underexposure in the dark parts -- now you can exposure the shadows correctly and know you'll be able to recover or tone down the highlights, which we can't do now. A simple banal example would be you can now take interior pictures and not worry about the blown out windows (although there will still be flare as you said).

[Rondec]

Yeah, in most cases I don't think it will be that big of a deal in terms of making "never before possible images" or something. Like I said before, it would be like a return to negative film -- right now we shoot on "direct positive", like slides. Of course we have much more leeway than with slides, but still blown highlights can never be recovered. Whereas with negative film, you didn't have to worry about the highlights so much (not unlimited, but 2-4 stops leeway) and could expose for the shadows. So this will be a welcome tech if they get it right (and there will be others ways to achieve it I'm sure). But it isn't revolutionary for standard use because we had it already for 100 years and we'd just be getting it back but with even more latitude. I'm sure you could also push the envelope and do some weird things with it, but that would involve long exposures and what not.

So for normal photography, this tech is not about avoiding overexposure as much as it is about avoiding underexposure in the dark parts -- now you can exposure the shadows correctly and know you'll be able to recover or tone down the highlights, which we can't do now. A simple banal example would be you can now take interior pictures and not worry about the blown out windows (although there will still be flare as you said).

Exactly. If I am taking a photo with the DA 15 limited, there are plenty of times at sunrise where I will try to expose for the sky and then bring up the shadows. But if I could expose the foreground without worrying as much about the sky, it would be easier to bring up the shadows without introducing a ton of noise.

[Nicolas06]

Yeah, in most cases I don't think it will be that big of a deal in terms of making "never before possible images" or something. Like I said before, it would be like a return to negative film -- right now we shoot on "direct positive", like slides. Of course we have much more leeway than with slides, but still blown highlights can never be recovered. Whereas with negative film, you didn't have to worry about the highlights so much (not unlimited, but 2-4 stops leeway) and could expose for the shadows. So this will be a welcome tech if they get it right (and there will be others ways to achieve it I'm sure). But it isn't revolutionary for standard use because we had it already for 100 years and we'd just be getting it back but with even more latitude. I'm sure you could also push the envelope and do some weird things with it, but that would involve long exposures and what not.

To me what count is the total dynamic range that I can get. There already at least some margin today on highlight on raws compared to JPEG. Under expose by 1-2EV and with raw and you don't need to worry of the highlight too much.

In fact the only case when I worry into highlight is for extreme cases (for example intérior with sun entering from windows) where the scene is more 20+ EV of dynamic range and where neither current sensor or film would keep details at all levels.

[Nicolas06]

Vs

That's typically an artistic choice. Well at least the first picture is better composed to me. But I guess the photographer wanted to make it not realistic. To me it look like it was send in B&W with a fax and then colorized !