Yep. In reality, there is precisely one degree of freedom for exposure. You have just one thing you can do - overexpose/underexpose compared to what your metering system tells you what the "correct" exposure is.
Anything you see on the back of the camera (histogram, flashing pixels for over/under exposure warnings, etc.) is just a guideline, not data to be fed into some hard-and-fast algorithm without thinking. Even back in the film days I used to eyball the scene first, then either set the exposure compensation dial on the camera or use manual exposure. Once I got a PZ-1p I could also use the pseudo-spot metering mode as a way of comparing different parts of the scene; on occasions I'd even use a hand-held meter (sometimes even using it in incident metering mode!). Do I get the exposure perfect every time? Probably not; but I'd guess that well over 99% of the time I'm off by no more than half a stop. And using a current-generation camera such as a K-5 or K-30 I rather doubt you could tell the difference between a theoretically perfect exposure and one under-exposed by half a stop (which I'm far more likely to end up with than one over-exposed by half a stop). And, if I'm in any doubt, I'm going to bracket; even without using the auto-bracketing ability of the camera I can take two (or three, or five, or however many I decide) frames, varying the exposure each time, faster than I can review a histogram. Sure, if you dig around in the RAW files, you might find that I've only got 12.5 bits of dynamic range, rather than the 13 a K-30 can deliver (or the 14 the K-5 seems to manage, according to DxO). But that's not going to matter unless you've got a display device capable of showing that much dynamic range, in an environment that allows it to do so. tl;dr Don't obsess over the histogram - look at the damn picture! On Wed, May 22, 2013 at 07:21:28PM -0500, Walt wrote: > It's amazing anyone ever took a decent digital photo before ETTR or > UniWB were devised, really. > > -- Walt > > > On 5/22/2013 5:41 PM, Ciprian Dorin Craciun wrote: > >On Thu, May 23, 2013 at 12:50 AM, Bruce Walker <[email protected]> > >wrote: > >>Now elsewhere you have explained that you want to doctor or calibrate > >>your histogram in aid of calculating exposures for doing ETTR. You > >>might want to consider that ETTR is considered by many to be no longer > >>relevant and even harmful. I don't follow the notion anymore myself. > >> > >>Have you read this? > >> > >>http://theonlinephotographer.typepad.com/the_online_photographer/2011/10/expose-to-the-right-is-a-bunch-of-bull.html > >> > >>Even doing nothing but RAW shooting I know that once you clip your > >>highlights, they are gone. Pure white. No recovery possible. Complete > >>loss of "value". Possibly still okay for showing to your parents. :-) > > > > Ok. I've quickly read the article, and as I understand it it boils > >down to this: > > > > (A) Since ETTR was introduced noise in shadows was reduced. (Thus > >there goes the initial motivation of ETTR.) > > > > (B) In case of high contrast scenes it is better to underexpose to > >catch a larger range of highlights instead of blowing them out to > >white. > > > > > > Now about (B) it makes a lot of sense in night scenes where there > >is artificial lightning mainly because the main subject of those > >photos are some lightened objects, thus we don't want to overexpose > >those. And as said previously I've learned this the hard way. > >(Moreover as the author says the same applies to other cases where we > >have high contrast.) > > > > > > But with (A) I tend to disagree somewhat... Although noise got > >better at lower ISO and / or in more expensive cameras, in "consumer" > >cameras like my K-30 the story is somewhat different: at ISO 800 the > >noise is perceptible, and at ISO 1600 it becomes bothersome in > >shadows. Thus if the scene permits me I would gladly try to apply ETTR > >to save some of that noise. (Of course if I get to higher ISO, it > >means my lightning is poor, thus most likely I must use a tripod or > >some place to put my camera onto.) > > > > Ciprian. > > > > > -- > PDML Pentax-Discuss Mail List > [email protected] > http://pdml.net/mailman/listinfo/pdml_pdml.net > to UNSUBSCRIBE from the PDML, please visit the link directly above and follow > the directions. -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
