"Live Histogram" on E-PL1

Started May 24, 2010 | Discussions thread
knickerhawk Veteran Member • Posts: 5,159
Re: Still doesn't solve the riddle

MikeK5117 wrote:

Doesn't matter, I don't know why you took that personally.

I didn't take it personally. I was just trying to assure you that you didn't need to educate me on the basics of exposure, the zone system, etc.

I have tried in more ways than I can imagine to explain to you that the two histograms are showing 2 different samples of the one scene, neither encompasses the whole scene. The histogram is only 9 stops wide, the scene may be 15 stops or more. Histogram #1 is of the shadow/central side of the overall scene, histogram #3 is of the highlight side (#2 is actually the furthest to the shadows). In histogram #3 you can see a spike of high value at the far right, in histogram #2 you can see a small hump of shadows that is on the far left, the total EV range represented bvetween those 2 points is much wider than 9 stops (I know this because in none of the 3 histograms do both of those datapoints show up at the same time), so that is why the histogram is different depending on which sample of the scene the meter tells it to represent. A histogram is not a picture, it is a sample of data, in this case it is a sub-sample of the total population represented in the scene.

Your description above of the three histograms as "samples" or "sub-samples" of the scene leads me to conclude that you're confusing histograms and metering and treating them as if they measurie the same thing. They don't. I've explained previously that histograms measure the lightness values of all pixels output by the camera either to the LCD in live mode or to a JPEG file. Histograms are constructed after the camera makes all sorts of adjustments to the initial meter reading, including exp. comp to shutter/aperture settings, demosaicing, gamma adjustment, WB adjustment, and other curve adjustments. Histograms come at the end of the exposure process, and metering is at the very beginning.

But now we're just going in circles, at least until you can answer the question of how two identically exposed images of the same scene, same lighting, same framing, same camera and same settings (other than metering mode used and exp. comp used to equalize the exposure) can have different histograms. My unshakeable conclusion is they can't unless one or both of the histograms is inaccurate or the type of histogram used is different (e.g., luminance vs RGB).

Happy shooting to you. I've enjoyed our repartee.
My photos: http://www.pbase.com/imageiseverything/root

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow