I think whether there appears to be enough shadow detail in the image is a
function of you having seen the scene when you shot it. Those who did not
are going to evaluate the image as it is, and I think it works nicely.
On Mon, May 11, 2015 at 2:52 PM, Tina Manley <tmanley@xxxxxxxxx> wrote:
> Thanks, Moose. The original dng does not look terribly underexposed. I
> can still see details in the shadows. Here it is:
>
> http://www.pbase.com/tinamanley/image/160035214
>
> Any more exposure and there would be even more totally blown out. I've
> always resisted using higher ISOs but I guess I'll try and see if it makes
> a difference.
>
> Tina
>
> On Mon, May 11, 2015 at 5:39 PM, Moose <olymoose@xxxxxxxxx> wrote:
>
> > On 5/10/2015 6:26 PM, Tina Manley wrote:
> >
> >> ...
> >>
> >> This particular photo was very dark. The ISO is only 320 but there is
> >> severe banding in the shadows. Why? It's there in the original dng
> >> without any manipulation. If the ISO is 320, just like most of the
> other
> >> B&W's in low light, why is there so much banding in this one?
> >>
> >> http://www.pbase.com/tinamanley/image/160019998
> >>
> >
> > Methinks you misunderstand the technology. There are certain combinations
> > of camera bodies and lenses that band because of RF interference, or some
> > such, but I don't think that's the problem here.
> >
> > This shot is deeply underexposed. I'm guessing from some of my own shots
> > that a good exposure for this scene is in the vicinity of 8 stops above
> > what you took. That's about what an averaging meter would give. That
> would,
> > of course, give a daylight like exposure, which is not what you want.
> > However, given a static subject, tripod and highlights that will be blown
> > either way, full exposure, then pulling the middle down in post, will
> > generally give the best results.
> >
> > You don't have that luxury with this subject. The only solution is higher
> > ISO and perhaps slightly slower shutter speed.
> >
> > This is an analog to digital device; it records analog input digitally,
> > which has a problem in A to D conversion. In the highlights, all is
> > wonderful. The top few stops have thousands of brightness values to
> record
> > what the sensor 'sees'. In the middle, there are still enough values to
> > exceed what our eyes can differentiate. At the bottom, there are fewer
> and
> > fewer values available. At the very bottom, there are only two values, 0
> > and 1, black and white.
> >
> > When you underexpose this much, you force the middle tones to be recorded
> > in a part of the histogram where there are few values and force the
> shadows
> > down into "insufficient information" land.
> >
> > You seem to think the banding (of this sort) is, or should be, an
> artifact
> > of high ISOs, when the opposite is the case. When you force the camera to
> > record everything in the bottom of the histogram, then when you pull it
> up
> > in post, there just isn't enough information to recreate the tonal detail
> > that was squished (technical term) in recording. When you use higher
> ISOs,
> > the camera amplifies the analog sensor signal before converting it to
> > digital. The result is generalized noise, but not banding.
> >
> > Bigger apertures and slower shutter speeds fix it all by delivering more
> > light to the sensor, but aren't available here.
> >
> > Another, broader way to look at it is that forcing anything, from
> material
> > to system to person, beyond it's limits will result is failure of some
> > sort. By underexposing, then trying to raise shadows well above any
> > actually useful level, you forced this particular system beyond its
> limits
> > into failure. It happens to be banding.
> >
> > If you want to find the practical limits, set up a test with static
> > subject, bracket exposures over a broad range and process them. You can
> > find the minimum exposure that gives the results you want at your desired
> > ISO.
> >
> > Does anybody know how to get rid of banding?
> >>
> >
> > 1. Underexpose less.
> >
> > 2. Since it's in areas that should be deeeep shadow on any print or web
> > display, don't worry about it, 'cause nobody can see it. Certainly no
> one,
> > including me, saw any problem with the image as posted.
> >
> > 3. NR can sometimes help. Blurring works pretty well in areas like sky
> > with no detail.
> >
> > Techie Moose
> >
> > --
> > What if the Hokey Pokey *IS* what it's all about?
> >
> > --
> > _________________________________________________________________
> > Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
> > Archives: http://lists.thomasclausen.net/mailman/private/olympus/
> > Themed Olympus Photo Exhibition: http://www.tope.nl/
> >
> >
> >
> --
> _________________________________________________________________
> Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
> Archives: http://lists.thomasclausen.net/mailman/private/olympus/
> Themed Olympus Photo Exhibition: http://www.tope.nl/
>
>
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|