10-bit Color

J

jduke

Member
I was testing my potential workflow in preparation for 10-bit external recording and found an unexpected effect:

* I have two 10-bit monitors
* Placed a 16-bit ramp on monitor #1. It looks clean, no banding visible
* Attached DVX200 with HDMI cable to monitor #2
* Set camera out to 4:2:2 10-bit 30 fps
* Pointed camera to monitor #1
* Monitor #2 shows clean 10-bit, no banding

But now:
* Changed camera out to 4:2:2 8-bit 60 fps
* Pointed camera to monitor #1
* Monitor #2 shows clean 10-bit, no banding

It is as if 4:2:2 8-bit 60 fps without recording is putting out 10-bit?!

If someone has an answer to what I am missing here, please let me know.

James Duke, jduke@lifestorypictures.com, http://www.lifestorypictures.com.
 
Your testing methodology assumes you can recognize 8bit banding by eye in your setup, do you have evidence to support this given?
 
You might be better off with something less subjective, for example shooting a gradient not on a screen so you know the subject is not a limiting factor, and then doing some equal post transformation (maybe increasing contrast to see the banding, just make sure you use a high bit depth transformation for appropriate comparison) to both images to stress the images and compare where they fall apart relative to each other. Keep in mind that in a camera noise may have a dithering effect and hide banding even in an 8bit camera image to some degree, and we look at 8bit and 6bit dithered to 8bit screens all day and don't necessarily see banding except in certain cases. For example you may want to use a lower contrast ramp or zoom into the ramp chart and also add contrast in post in an equal amount for both images. There is a reason 8bit is so widely used for delivery, it isn't as bad as some think if it's handled and processed correctly, it just doesn't stand up so well to post.
 
You must log in or register to reply here.
Back
Top