How do I use the live histogram? I know a spike to the left means

>>2987042
>I know a spike to the left means blown out darks and to the right means blown out highlights
This suggests you don't know what the histogram actually is. Google that. You'll find a guide that will probably also explain a few use cases like the ones you gave

>>2987042
Did you learn how to read a X Y graph at school?

Y is the intensity value

X is a spectrum from dark to light

>>2987082
how do i read swahili you stupid fuck

>>2987154
huh

>>2987154
>swahili means "incomprehensible dialect" for ameridumbs

wow, such retard.

>>2987154
Kek, you're just a retarded person. First finish high school before posting here.

OP here. Clarifying: i'm not asking how to read a histogram, but rather what to make with that info in specific situations (high contrast, very dark ambients with strong highlights, etc)

>>2987046
I couldn't find any thorough tutorial on this subject

>>2987206
>I'm nothing asking how to read
>I'm asking what to do with the info

Top kek american education.

>>2987213
xD

Post a BW picture and I will explain it to you.

>>2987206

You interpret by figuring out how you want your photo to look in the first place. Shooting something bright and white, like snow? Try to increase exposure until the histogram is on the white side.

Shooting something dark? Try to bunch it up on the left side without clipping.

Contrasty situation? Not much you can do. Keep an eye on the right side of the histogram to make sure you aren't clipping highlights, as shadows are much easier to bring up than highlights are to bring down.

Basically, you do the driving, and the histogram is just your speedometer.

>>2987230

Oh, and I forgot probably the most practical application for histograms: keep it in the middle most of the time if you aren't sure whether you exposed correctly or not. Middle means a well-exposed image most of the time. I was underexposing things for years out of a fear of using too high an iso, too slow a shutter speed, foolishly using the lcd to judge exposure, and then lacking a critical eye to discern whether an image was underexposed or not.

Using a histogram and just bringing it up to the middle will, a lot of the time, result in a good exposure.

File: 1481560000292.png (15KB, 1310x337px) Image search: [Google]

15KB, 1310x337px

>>2987042
pic related
also learn to use the infographics thread >>2981768

>>2987272
>high contrast
That's actually wrong. There's no pure black or white to be considered high contrast. Also, it has to much gray

>>2987341
yeah, it's more like high dynamic range and low dynamic range

Relatedly: if my camera's live histogram tells me that all the stuff in frame will be in the center of the histogram, and my camera's sensor exhibits ISO invariance, is it OK to EC down until the histogram just barely touches the left edge? Assuming that my intent is to get a stop or two more shutter speed.

>>2987384

yeah, that's fine. it's functionally the same as raising the ISO.

>>2987384
live histograms for most cameras are based on the compressed JPEG engine, rather than the actual RAW data from sensor.

This means that the left edge and right edge of the histogram being displayed on the camera are unreliable as hell.

If trying to edge the histogram, do a manual HDR, take one picture a little bit beyond the edge, at the edge and pulled back from the edge.
Sometimes the photo will have a larger usable range than the histogram jpeg displayed.
Sometimes it might be clipping and the jpeg is lying bullshit.

I hope canikon fixes that shit.

>>2987384
>overexposure warning are based on the JPEG-preview, thus the raw data undergoes certain transformations: white-balancing, tonal curve application, and also the color saturation is adjusted. These transformations depend on the camera settings for white balance, contrast, saturation, color space, sharpening, etc; thus for the same scene if the camera settings are changed the histogram is also displayed differently. At the same time, the raw data depends on the exposure and the selected ISO speed rating, while other camera settings have no impact on the raw data. In other words, for a raw shooter, the histogram does not accurately reflect the real data in the image file.

There are tests where the histogram is 0.5-1 stop too cautious (could've gone a stop further before overexposing) and where it's doesn't even complain when most of the entire image is blown out. So I would most definitely not choose your workflow based off of simply hugging left or right side of histogram jpeg.
http://www.rawdigger.com/howtouse/beware-histogram

>>2987206
good luck making it in life as a beta human

Live histogram shooting is most useful for ooc jpegs, and actually makes it very simple to get a perfect exposure on the first shot. Which saves clutter on your phone if you do a wireless transfer.

For raw, my camera's auto exposure is good enough that I never worry about clipping or excessive underexposure. If im serious about image quality, I'm bracketing anyways.

Why don't you just turn off all those stupid overlays so you can actually see the photo you're trying to take? Can't you just judge the exposure with your fucking eyes?

>>2990359
Actually no, and lcd screen in daylight is extremely difficult to judge. Some EVFs are easy but not all, especially if the sun is hitting the corner of your eye.

I do agree that's way too much shit in OP's pic. One histogram is all I need

If I looked at histograms before I took pictures I'd never take any, the scene would have moved on most of the time. I can shoot a couple of frames and then bracket at +2, +4 and -2, -4 in the time it takes to review a histogram. In reality experience tells me which way I need to compensate anyway so I never even shoot that much.

I bought a big, fucking expensive camera I trust it to make some of the decisions while I concentrate on the shot

>>2990368
>one histogram is all I need
to produce your shit-tier talentless snaps, sure.

Actual non-trash photographers at minimum switch to RGB histogram whenever a scene is very colorful, as luminosity does not have the necessary info to know when one of the three colors clipped and lost all texture/gradation when the other two colors weren't clipped. Luminosity hides the story entirely.