Thread: Correct Exposure on the using false color

Results 1 to 6 of 6
  1. #1 Correct Exposure on the using false color 
    Hey all - been a long time lurker. I have a few questions about proper exposure on the Ursa Mini 4.6k, especially when using false color. I come from a stills background, so it took a while to acclimate myself with the "ETTR" method. I have tried a few case of this recently, and found that my colors end up washed out and the footage isn't all the pretty.


    Just a few prefaces - I am NOT looking for "it all comes down to creative decision, man!" type of answers. I am fully aware of this and I am simply attempting to understand the foundation of how to properly expose for a clean image in this camera! I only hope that I can cultivate at thread where there are a lot of informative answers that myself and hopefully others in the future can use.

    ------------

    Scenario 1:
    Outside, with the subject lit by natural light. Focus on skintones with the assumption that nothing in the background is going to be blown out, and if it is, it doesn't matter anyway.


    When in instances where there is no lightmeter, false color seems to be pretty handy. However the BM false color is balls compared to something like the smallHD in my opinion. Right now I understand that skin color should be pink which sits roughly in the 48 IRE range. The "sweet spot" for ETTR is 1.5-2 stops on the UM 4.6k. What colors should I be looking for in that case? The pink area is already 1 stop over middle gray, am I supposed to push it 1 more stop (making it 2 stops over middle gray) or a full 2 stops above the pink region?

    ----------------
    Scenario 2:

    You're shooting a dark scene using ambient street light. (let's forget about white balance issues for now) Where should your skin tones land in order for you to start over exposing to keep a moody shot as noise free as possible? Where are they supposed to land after pushing the image? The same question applies if I am in a controlled setting outside, and I set my lighting ratios the way I want them. I know I want a darker image for atmosphere, but I am unsure how I am supposed to light my subjects and by that I mean giving the sensor enough light to capture the information needed but also not lighting them so they are lit as if they were out in daylight.

    I know in *this* scenario there are no hard formula answers but I would love to know where to at least START so I can start knowing what to look for.

    ---------------
    Scenario 3:

    This is less of a scenario but a question regarding lighting at ISO 400, then increasing the ISO to 800 effectively giving it one more stop. If I want my shot at 2.8, would it be more effective to close down the Iris one or two Stops - THEN adjust my lights and ratios - and then finally open back up to 2.8? In both cases you are saturating the sensor with more light because you are rating the camera 1-2 stops under what you're shooting. I ask this because the native ISO is 800 in the BM. -

    Finally, assuming this is all in Raw - once you're in Davinci Resolve is it as easy as using the exposure dial to bring back the image you were seeing before over exposing?


    My workflow as I understand it right now/what I've been doing:
    I basically shoot/light at 400. Then once I get an image I like using a LUT preview on the LCD screen, I bring up to 800. In theory (at least I think) I'm giving myself one stop of latitude in post.


    I know it's a lot, but I've seen so much gorgeous footage from this camera that it's had me scratching my head as to why everything I've been shooting doesn't really come close. I remember shooting on the Pocket and the overexposing until zebras had served me well there. Even Franks instant cinema lut as been giving me mixed results (it seems to bring down the exposure by at least a stop once applying it). Finally, if anyone has any sceenshot examples of how they captured there images using false color along with the final image, I and many other people would be eternally grateful!!
    Reply With Quote  
     

  2. #2  
    Senior Member
    Join Date
    Apr 2016
    Posts
    102
    I think it's easy to overthink ETTR. The key idea is to get away from the noise floor, to collect more light than you plan to eventually use.

    So, if you're shooting raw on a camera with native ISO of 800, rate the camera at 400 or 200, adjust lighting and aperture until you like what you see in the monitor, don't clip highlights, and you're good.

    No need to change ISO back to 800 only to pull it back in Resolve. If you're shooting raw all you're doing is going in a circle. ProRes is a different story since in that case you're baking in the ISO setting.
    Reply With Quote  
     

  3. #3  
    Senior Member
    Join Date
    Jan 2014
    Location
    Los Angeles
    Posts
    129
    As frustrating at it might feel you're doing the work and I really appreciate the amount of effort you're putting into your images. why don't you post some screen grabs from your shoots and ask for feedback. I'd love to see what your doing. the only way that I've been able to figure out these questions you're asking is to do camera tests myself. see what works for me. I think that's why people are hesitant to just throw out numbers.
    Reply With Quote  
     

  4. #4  
    I.

    I think that the term Global Warming is a much less accurate, much less effective term than Climate Change. In the same way, I think ETTR is a much less effective way of saying that you should expose your images in such a way that you capture critical information without letting it clip. Whether that means exposing for higher ISOs to protect highlight information or exposing for lower ISOs to protect shadow information or decrease the perception of noise, you should always be optimizing your use of your sensor's dynamic range on a scene by scene basis.

    II.

    Just because nothing is clipping doesn't mean that everything is captured accurately. When a camera is exposed almost to the point of complete saturation, getting a good image from it isn't always as simple as adjusting the ISO in a drop down menu. If you expose for lower ISOs you're going to get less noise but it's also very likely that you can lose a bit of precision and saturation. Try exposing with analog tools (light metre) at ISO 800 and see what that gets you. I think you'll be pleasantly surprised by the results.

    III.

    A lot of what makes an image pop is from how you manipulate the image in post production. Very seldom do I get a frame straight out of camera and think that it is perfect just the way it is. Everyone has their own recipe, everyone has their own look. Wether that's noise removal, vignettes, re-framing, artificial noise, bespoke LUTs, power windows, HSL, etc. is up to you; just remember that there's a lot of work that happens after you hit record too.
    Reply With Quote  
     

  5. #5  
    Quote Originally Posted by Alex.Mitchell View Post
    I.

    I think that the term Global Warming is a much less accurate, much less effective term than Climate Change. In the same way, I think ETTR is a much less effective way of saying that you should expose your images in such a way that you capture critical information without letting it clip. Whether that means exposing for higher ISOs to protect highlight information or exposing for lower ISOs to protect shadow information or decrease the perception of noise, you should always be optimizing your use of your sensor's dynamic range on a scene by scene basis.

    II.

    Just because nothing is clipping doesn't mean that everything is captured accurately. When a camera is exposed almost to the point of complete saturation, getting a good image from it isn't always as simple as adjusting the ISO in a drop down menu. If you expose for lower ISOs you're going to get less noise but it's also very likely that you can lose a bit of precision and saturation. Try exposing with analog tools (light metre) at ISO 800 and see what that gets you. I think you'll be pleasantly surprised by the results.

    III.

    A lot of what makes an image pop is from how you manipulate the image in post production. Very seldom do I get a frame straight out of camera and think that it is perfect just the way it is. Everyone has their own recipe, everyone has their own look. Wether that's noise removal, vignettes, re-framing, artificial noise, bespoke LUTs, power windows, HSL, etc. is up to you; just remember that there's a lot of work that happens after you hit record too.
    While I appreciate you taking the time to answer, I find it a little puzzling. Maybe I wasn't clear - I understand the work that goes into post in order to make an image *pop*, it is the same in photography. However, in photography I can take a picture (albiet maybe underexposing by 1/3rd a stop) and make very simple adjustments to saturation and contrast to come out with a very usable image. Maybe I should rephrase my question:

    When using false color, focusing only on caucasian skintone and ignoring any creative choices that affect the "mood" on a cloudy afternoon - what exactly am I looking for? Is the face supposed to be mostly pink with green highlights? Is it supposed to be mostly gray with pink highlights?

    AND

    How does this change with overexposing? If I am making use of the cameras dynamic range for the entire scene, how far can I push the skin and retain all the detail/color/saturation?

    Hopefully this is much clearer.

    As frustrating at it might feel you're doing the work and I really appreciate the amount of effort you're putting into your images. why don't you post some screen grabs from your shoots and ask for feedback. I'd love to see what your doing. the only way that I've been able to figure out these questions you're asking is to do camera tests myself. see what works for me. I think that's why people are hesitant to just throw out numbers.
    On it! Weather has been less than stellar this past week but I should be out in the field soon.
    Reply With Quote  
     

  6. #6  
    Senior Member
    Join Date
    Jun 2013
    Posts
    312
    Regarding Scenario 2/3: this is not specific to false color, but Frank had a great post on how he exposed for similar scenarios. He "overlit" by 2 stops: https://frankglencairn.wordpress.com...ergiversation/
    Reply With Quote  
     

Similar Threads

  1. False Color Monitor
    By John Fulp in forum General Discussion
    Replies: 9
    Last Post: 08-21-2016, 08:13 AM
  2. Replies: 5
    Last Post: 04-05-2016, 12:24 AM
  3. Where did you learn to color correct?
    By bigwidge in forum Workflow
    Replies: 15
    Last Post: 02-09-2015, 12:44 PM
  4. False color
    By Matt_Rozier in forum General Discussion
    Replies: 8
    Last Post: 03-17-2014, 06:47 PM
  5. Replies: 7
    Last Post: 10-06-2013, 11:23 AM
Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •