Brian Swale wrote:
>Hi all,
>
>AG (and others) have written knowledgeably about the precise range of
>exposure latitude possessed by an E-series camera. From my experience I
>also agree that it is more akin to slide film than print film, and especially
>regarding landscapes, or other shots with bright sky in the view-finder, the
>sky usually will get blown out. With print film anyway, that is less likely to
>happen.
>
>
Brian Swale also wrote:
>There are a lot of things that E- cameras struggle with and or fail. For
>example, you may note that many photographers avoid bright sky in shots
>were other subject matter is a bit dark. Why? Because the latitude of the E-
>cameras is rather small. Print film has probably a wider latitude.
>
Brian, your questions and coments on E-thing latitude generated some
interesting and useful posts in response. However, there is one
important facet that wasn't commented on:
Previously, Brian Swale wrote:
> I am using SHQ jpegs. More high quality shots per CF 256MB card.
>
> I'm not sure that Viewer would do that on my machine(s); I run it on
> my lap-
> top with W'98, and it runs with only some functions working; ask too much
> and it crashes.
If you want to talk about the exposure latitude of any digicam that will
produce RAW or TIFF image files, you cannot use JPEGs for meaningful
evaluation. JPEGs do two kinds of compression for these cameras. The one
we are concerned with here is in brightness information. JPEGs are
limited to 8 bits of brightness data per color per pixel. Most DSLRs
produce 12 bits. (The E-1 produces 14 bits, but it doesn't appear to
actually output any greater latitude than others.)
So you have a situation a bit like hooking up your garden hose to a fire
hydrant or a water main. No matter how much the source is capable of
delivering, no more comes out of the hose. Your E-1 is capable of
producing quite a broad latitude. But when it outputs JPEGs, it has to
figure out how to compress 4,096 (12 bits) or 16,384 (14 bits) of
brightness values into the 256 values that 8 bits in a JPEG allows.
Different manufactures use different strategies and you can modify them
somewhat with the image parameter settings of the camera. The result is
generally a combination of modest compression of central values,
disgarding of the very extremes and more extensive compression of the
remaining highlight and shadow values (creating the toe and shoulders).
In practice, JPEGs tend to have about the same latitude as slide film,
as you have discovered, but that is a limitation of the image format,
not the camera. I don't know the latitude of the E-1, but guess it is at
least 6 stops, and likely more.
You make clear the reasons why you shoot JPEGs, and that's fine.
However, it isn't right to complain about the camera's "struggle with
and fail" problem with latitude when you are intentionally limiting it
yourself to conserve storage space and avoid computer upgrade expenses.
ALL DSLRs are capable of much greater exposure latitude than you are
getting by shooting JPEGs. There are of course many shooting situations
where all that latitude simple isn't needed, and JPEGs are just fine for
those.
The question that I don't know the answer to is why there isn't a JPEG12
image format. The amount of extra programming effort to carry 12 bits,
rather than 8 or 16, should be trivial.
Moose
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|