Histogram is not the same in camera vs. external monitor

Started by albertgarciarue, July 12, 2015, 07:17:05 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

albertgarciarue

Hi, I'm using a 5D Mark III with an Ikan VK7i via HDMI and shooting H.264 for this tests.

I used to trust the Magic Lantern histogram when using an external monitor until I realized the histogram was not real: it underexposes the histogram by 2-3 stops, again, the histogram is underexposed, not the image, which then is overexposed by 2-3 stops.
Is that normal? What is causing the problem? For this specific test I was not using the RAW histogram since I'm shooting H.264.

Here's a quick test. The image is actually not showing in the screenshot, but the scene was even except for a white cloth, which size was about 20% of the scene.
- Screen capture when using the monitor. Pretty good exposure, huh? Looking at the image (which is not showing in the screenshot), the white cloth looked overexposed, but the histogram didn't say that. I trusted the histogram to set the exposure.


- Screen capture when using only the camera: I simply turned off the monitor, I didn't changed any exposure values, as you can see. This is the actual image that would be recorded (the image is not showing in the screenshot, just look at the histogram). It's clearly overexposed by 2-3 stops, and the white cloth is clearly blown (I tell you that).


Then I guess I have to set the exposure using the camera and then turn on the monitor and don't look at that false histogram again.

Any tips?