Old single Raw pictures processed HDR

[musicaleCA]

Camera sensors have dynamic ranges around 12 stops at the best of times (a very high quality sensor at ISO 100). The human eye has around 20. HDRI approaches or exceeds this 20-stop dynamic range, and then is tone-mapped so that it can actually be displayed on current media. A single RAW file cannot produce an HDR image, period. And furthermore, an actual HDR image will produce better detail in the shadows than a single RAW exposure, every time.

 

[ccarollo]

Raw files have between 12 and 14 bits of precision. A jpg has 8 bits. A series of bracketed raw files has up to 32 bits of precision, depending on how well you've bracketed the scene.

So yeah, tonemapping a raw file down to 8 bits is a reasonable thing to do. It's just not a whole lot of dynamic range to start with, so it won't work with any scene with a lot of dynamic range.

 

[Buckster]

No need to feel sorry sweety.

Is it your intention to purposely piss people off? I dont speak for NateS, but Im offended by your alteration of his name from NateS to sweety.

It's a term of endearment, just to lighten the mood a little. Heck, we're just shootin' the breeze here with opinions anyway, right? No need to take any of it personally, and certainly no need to be offended over any of it. Say "CHEESE"! :mrgreen:

I just see things a bit differently maybe. As mentioned a couple times now, if the scene doesn't have too wide a range to begin with, three acceptable frames that deliver the detail necessary for an HDR combo to show them all is possible.

And when we look at the end result, if we're not told the method, it would be nigh impossible to discern, in most cases, whether the end result image we're looking at was made with three shutter actuations or three RAW outputs, so what's the real difference?

I'm also going to throw this out there... Most of you would agree that a well-made HDR image is not one that looks 'cartoonish' but, instead, looks acceptably 'real'. And how is that achieved? By not going overboard with it. That means we let some of the darkest shadow areas stay very dark, and we let some of the brightest areas stay very bright. I'm not talking photographic black or photographic white - totally blown out. I'm just talking about a reasonable degree of contrast remaining in the image that allows our brains to see it as 'natural' looking.

Well, if that's the case, we rarely need or want or should use all the possible dynamic range that is available. You know the old saying, "just because you can, doesn't mean you should". In the same way, just because it's available doesn't mean we should necessarily be using it. We're not trying to make everything in the scene become 18% gray in the resulting image, after all - that's how we get cartoonish mud.

So, again, I think this comes down to a perception issue between the viewer and the end result image - not necessarily how it was achieved.

It's like looking at a B&W photo in someone's hand. If we're not told, we don't know if it was captured with color that was discarded in post or if it was made as a B&W at the moment the shutter snapped. Why should any of us rear up like the B&W police and say, "that's not a real B&W", when what we're looking at is, indeed, a B&W image - no matter how it got there?

Again, that's just how I see things. No sweat if others don't. :thumbup:

 

[ccarollo]

And when we look at the end result, if we're not told the method, it would be nigh impossible to discern, in most cases, whether the end result image we're looking at was made with three shutter actuations or three RAW outputs, so what's the real difference?

Given the limited dynamic range of a single RAW image, I'd wager that that's typically not true. It's pretty rare that tonemapping down a RAW image is going to look nearly as good, or natural, as a bracketed series of exposures. In certain scenes, absolutely, but it's the exception rather than the rule.

I agree with your statement about making HDR images look natural, but at the same time, if you're not capturing all of the possible dynamic range, than you will necessarily have photographic white or photographic black. Which may be fine, depending on what you're going for. But you can't not capture the full range of a scene and at the same time avoid clipped whites/blacks.

 

[Aeb]

I must say this thread contains such a wealth of in-depth information on the subject of HDR by the participants that will absolutely benefit not just any HDR fan, but I would say all digital photographers. In particular, the disagreements put forth are highly beneficial and they provide even greater insight to all on the subject.

Whilst it is true that working with the content of RAW files can be technically more satisfying than working with 'flattened' JPEGs, from the point of view of getting the final desired result, it may however be just be a matter of using different ways of skinning the cat.

The existence of RAW digital storage formats is such a wonderful development in the history of photography that it literally changed the world as see it. And with further advancement in optical sensor technology which will enable even more data information to be digitally recordable via suitable storage formats, the future is sure to be more exciting.

If one may use a sound analogy, one may say that a RAW digital file is able to provide 'timbre' and 'harmonics' playback, which JPEGS or other similar 'flattened' formats can't, regardless of bit depth used.