False Color Pipeline

[gx42]

I can’t remember if this has been asked here before, but I figured I would re-visit in a different light.

I respect that Roger puts an emphasis on exposing by eye, and at times also by meter. I think we should keep that in mind and work to learn it with practice. Yet practically, until we are at that point, having tools like false color really helps if you’re questioning some light levels on the day. I think it can also be a tool to expedite the learning process on exposing by eye, and also find the limitations of your camera’s dynamic range, giving you feedback right there on set. This can also be useful if you’re relying on a monitor that perhaps wasn’t calibrated, these things happen.

I say all this just to recognize that false color, like many other things, is just a tool. Something that can be useful, but presents its own issues and complications.

My question involves the dilemma between pipelines. You have EL Zone false color, which is reading the LOG or RAW image data and providing feedback on that. And then you have Arri False Color, which can read the image with your basic monitoring LUT applied, and provides lighting feedback on an image that is crushed a bit into rec709, etc.

Aside from shooting I have also used it to study. Sometimes I’ll plug my SmallHD into the laptop and turned on false color to see what some ratios look like in a scene that caught my attention. This is obviously reading the levels of a converted and graded image, not a log or raw file. I guess you could also do this with screen grabs and importing to Resolve, same idea.

This is kind of an open ended question on workflow preference, accuracy, and potential drawbacks. Is there a correct or preferred pipeline? And drawbacks of one versus the other? I’m curious what Roger and David think of of these 2 different approaches, even if they don’t use it themselves.

Thank you!

 

 

 

Topic

[trevorpendleton]

Ditto I would love more insight into this. I use a basic rec 709 transform and monitor the false color through that. My problem is personally I want the cleanest image possible to go to post. Least amount of noise possible. It would be helpful to have tools or a process to double check to make sure I’m feeding the sensor enough light. I know some people use ETTR but I don’t believe Deakins uses this and his images are so clean. Even in dark silouhetted scenes, there’s no noise, it has so much information still there in the darker parts of the image, for example in Skyfall in the church scene. Usually there’s also bright highlights as well, but also the darker parts are so healthy with information. It feels dark while also giving the camera clearly enough photons to be perfectly clean (and feel rich). That’s something I want to learn how to do, and a way to think about doing that.

[Author]

Replies