Thread: Noisy Shots - Am I doing something wrong?

Reply to Thread
Page 2 of 2 FirstFirst 12
Results 11 to 16 of 16
  1. #11  
    Senior Member
    Join Date
    Dec 2014
    Location
    Sarajevo
    Posts
    791
    DOWNLOAD!

    Hm, can't see much noise. DEB and ADD used...
    Reply With Quote  
     

  2. #12  
    Senior Member Zeb B's Avatar
    Join Date
    Sep 2011
    Location
    In The Moment
    Posts
    4,932
    Light to just below triggering the Stop Lights
    Shoot at lowest RC Compression possible
    Learn to love the "organic texture" Dragon provides
    or
    Convert it to B&W and call it artistic grain ;-)

    • Seriously, nothing in the couple in living room that stuck out camera wise other than it's underexposed by a couple stops if high key is what you're going for.
    • Any time you have to Lift Dragon footage in post . . . you're going to have a bad time, if low noise is your goal. Dragon is a light piggy. Feed it
    • I would probably light it differently (bounce off ceiling = meh Real Estate photography). Time of day? Warm it up. Add some haze to soften everything. Bring "sun light" through the blinds via big ass light plus gel. Not sure what the art direction was tho.
    • Perhaps use a more natural fake preggers belly on the model (the pillow looks like . . . a pillow)
    • Have some camera movement to get some energy into the shot. The wonky parallax of the static verticals is distracting.
    • In the edit I would use this specific shot for about 1-2 seconds with a slight zoom push to follow him. Then cut to her lifting her head and facial reaction of love for him, reverse shot of him coming in, his proud face, gimbal shot of his strong hand on the baby carrier, back to her face, then CU of her hand (with wedding band of course) rubbing her future pillow baby, his hand gently covering her hand, CU of their eyes locked on each other, fade to product ;-)
    -Zeb
    FAA 333 Drone Ace
    6K Drone Worldwide
    Underwater Shooter
    RED in Hawaii w/Operator
    www.Stock8K.com
    Reply With Quote  
     

  3. #13  
    Thanks for all the input guys...

    If 800ISO is not correct then I definitely am underexposing many shots and is a good place to move forward. I remember back in the day using Fuji Velvia rated at ISO 50, to truly it was a 40. Only a 1/3rd diff, but slide film has little to no lattitude.
    Zeb... regarding your comments about fake pregnant pillows, and verticle paralax and the rest... this was last shot of the day as an "add-in" and the way this will be edited with a whole bunch of other clips that will be heavily graded those issues won't matter. You might only see one second of this. I just thought it a good example to ask about the pixelation. And also as a comparison to the example demo reel which used the same camera, and yet was super clean. Granted this is raw ungraded vs. polished to the Nth degree. Still I have many other clips where I'm still seeing this. I think shooting at 400ISO is a good place to start with this though and I'll see what that gets me.
    Reply With Quote  
     

  4. #14  
    Senior Member
    Join Date
    Dec 2010
    Location
    Toronto & Vancouver
    Posts
    3,462
    Just an FYI, next time make the clips smaller, as even 5seconds is enough to tell (it's not necessary to have a couple gigabyte clips when ~300meg clips will do).

    As other have said, keep your ISO on the conservative side (320~640) as that will force you to use more light and/or open up your physical exposure (aperture and shutter). Also depends on what OLPF you're using. I think the general consensus is Skintone 250~500, Standard 500~800, and LLO 800~1600 (using the lower side with tungsten and the higher side with daylight)

    Sure ISO is metadata, but it doesn't work the way you'd think; the image that the camera records is always the same "sensitivity" and the ISO you apply to said image data afterward merely amplifies whatever sensor data was recorded (aka amplifying noise if there wasn't enough light). Thats why setting your ISO is still kind of important with RED, since that will dictate how you expose (and unfortunately REDs, even Dragon and newer, aren't quite as forgiving as Alexa/ALEV is with over/under exposure.)

    That said, I looked at the clip of the girl on the laptop and it looks acceptably clean on my Retina display... Like as clean as the reel you linked too. Are you sure you're looking at/rendering your footage at full debayer? One thing I've also always noticed is that whatever recompression Vimeo/youtube does almost always reduces the amount of visible noise (even film grain applied in post often gets mitigated by whatever recompression they do). So are far as that reel you linked to, I bet that the Vimeo recompression makes it look cleaner than the full-resolution raw (and I still think your footage is comparably clean).

    OH, another thing, try to stick to daylight balanced sources. I know Dragon and Helium are much better than MX was with non-daylight, but I'm always surprised with how much better daylight looks (especially in low-light when that pesky blue-channel will be struggling).
    Reply With Quote  
     

  5. #15  
    WOW... yeah good stuff to know. My first experience with this whole Red ecosystem was a couple years ago when I went to a production agency that was selling their Epic-MX. The guy showing me the camera is saying "Yeah we don't worry about ISO... we just set it and forget it 'cause we know we can change it in post." That's not what I've been doing, but it sort of set the tone of "wow it's that flexible..." Needless to say, I opted to NOT buy the camera when I watched this guy just chuck the redmag about 6-10 feet across the room into the case and knock around a bit. I shoot this next Sunday at the latest... will take all of this into account!
    Reply With Quote  
     

  6. #16  
    Member Shaeden Gallegos's Avatar
    Join Date
    Mar 2014
    Location
    LA
    Posts
    90
    Also, this is stated a lot, but RED doesn't apply any noise reduction in-camera like a lot of other manufacturers do. So if you try to compare to another camera that has applied an algorithm, it isn't a fair comparison. You can try and add some reduction in Resolve or a plug-in like "Neat Video" that does wonders :) But the other tips are great too; try lowering compression and ISO a bit.
    You don't have to see something to believe in it, but you have to believe in something to see it.
    Reply With Quote  
     

Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts