Dan Mitchell wrote:
> Moose wrote:
>
>>> One thing that confuses me is that an "HDR" image actually has a lower
>>> dynamic range than the original scene did
>>>
>> Not in the 32 bit mode used by some HDR apps. I'm too lazy to do the
>> research and math, but I think 32 bits may be enough to encompass the range
>> from direct sun, as seen on a clear day through the atmosphere, to below the
>> lowest threshold of light for the adapted eye.
>>
> Well, 32 bits just gives you more resolution, but surely dynamic range is the
> difference between 0 and the top, not the number of steps in the middle. 1
> bit would be enough if you're really not picky, 8 bits is certainly enough to
> represent a scale between total blackness and "the sun up close" -- there
> just wouldn't be a lot of detail available in the 'interesting' range.
>
C'mon Dan, you know we are talking here about range to contain the
combined DR of several 12-14 bit, bracketed shots. I mean it can hold
all that without losing any of the tonal detail of the originals.
>> So the combined capture can be true HDR - in a file. What you are dealing
>> with is the lack of display technology to actually display anything like the
>> full range of brightness.
>>
> Exactly -- if monitors were a:big enough to force the eye to open/close to
> compensate for varying brightness, and b:better at subtle shades of darkness
> and able to be much brighter, then this could be close -- and at that point,
> you'd want to have a lot of resolution to take advantage of that extra range
> of brightness.
>
Agreed. I've never seen IMAX, but the opening of Lawrence of Arabia on
the old, three camera/projector Cinerama had that quality. For viewing
in an ordinary house, I'm not sure I really want that. Even if my 55" TV
could do Sun level brightness, I don't think I would want it to.
>> Even slide film could hold more DR than most photo paper can display.
>>
> And depending on how opaque unexposed slide film is, and how bright the bulb
> in the projector is, you could possibly get close to the real world.
>
Nah, slide film has way too narrow a DR, much less than CN.
>> From fairly early on in the development of film to now, one of the skills of
>> a good photographer
>> has been to compress and/or clip that range of tones into a form that reads
>> as realistic to the human eye. That process is, for example, the basic
>> reason for being of the Zone System.
>>
> My point exactly == "compress and/or clip" is reducing the dynamic range of a
> scene to a range that works for display on the medium in question, be that a
> monitor or photo paper.
>
And yet, I object to the characterization of HDR, "an "HDR" image
actually has a lower dynamic range than the original scene did"
based on the limited output DR, without reference to the much greater
range in the file.
> HDR images with tone compression and the like are just doing this with a
> larger original dynamic range than has been previously possible.
You say "just". I think it rather wonderful to have the full range to
work with in making my own compression and clipping decisions. I can do
more than I could with a more limited original file. Even the greater
range of a 5D RAW file has much shadow/highlight detail to work with
than slide film.
> (hm. I suppose with judicious masking and bracketed negatives, you could do
> the same thing in a darkroom, it would just be awkward).
>
You can bet it's been done.
Moose
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|