Thread: 8k for stabilisation

Reply to Thread
Results 1 to 8 of 8
  1. #1 8k for stabilisation 
    Hey everyone,

    Wondering if anyone can help with this.

    I'm shooting a project on my Helium 8k in 8k WS and I'm going to want to stabilize in post. I'm looking for the best way to achieve this. I've heard of the technique of framing for 4k in the monitor and then scaling in in post to find that "framed for 4k" shot value and then stabilizing.

    So, utilizing this technique, is any normal stabilization within commonly used NLEs (like stabilization within fcpx or warp stabilizer in Premiere) the best way to stabilize given all that extra room? Or is there another, better way? Also, how do I frame for 4k WS (or any other lower resolution than 8k WS) in the monitor? Couldn't seem to find how to do that.

    And also, how do I get a mathematically perfect match for the 4k WS frame in post when I drop the clip in the timeline? Do I just approximate that frame by scaling in until it looks correct?

    Thanks for the help!
    Reply With Quote  
     

  2. #2  
    Hi Tom. If your final product is 4K, I wouldn't frame for 4K. Rather use as much of the sensor as possible while leaving some room for stabilisation. How much depends on how much you will need to stabilise in post. Also bear in mind that you will need to shoot at relatively fast shutter speed to avoid too much motion blur.
    Reply With Quote  
     

  3. #3  
    Hi Robert, thank you for the reply and for the shutter speed tip. Hadn't thought of that. I imagine it's to help smooth out the motion blur created by the stabilization?

    For this shoot, I'll be pretty steady. I'm shooting on a Ronin 1 with easyrig support. I just want the shot to be steadycam smooth. The Ronin will be nice and stable but I still want to leave myself with some more options for stabilization should the Ronin move leave something to be desired.

    So if I were to frame slightly within my monitor, what type of frame guide would you recommend? Or should I just eyeball it given the amount of stabilization I'll feel I'll need in post? Is it possible to frame for a certain resolution?
    Reply With Quote  
     

  4. #4  
    Senior Member Aaron Lochert's Avatar
    Join Date
    Oct 2013
    Location
    Tucson, AZ
    Posts
    938
    Quote Originally Posted by Tom McNamara View Post
    Hi Robert, thank you for the reply and for the shutter speed tip. Hadn't thought of that. I imagine it's to help smooth out the motion blur created by the stabilization?
    Not only that, but when motion blur is present, stabilization is harder to accomplish. If every frame is sharp, you'll get better tracking. On the other hand, depending how fast your camera is moving through a scene, a lack of motion blur can make things stuttery and counteract the smoothness you want to feel in the shot. Ideally you find a happy medium for this in testing. But since you said you'll be on a Ronin, I don't imagine any hard bumps that will throw off the stabilization. I would first test a 180 degree shutter and see if you need to go for something shorter.

    I would definitely frame for 6K and not 4K. Stepping down to 4K is a 200% punch in and the quality will definitely suffer in comparison to shots in the project where you aren't punching in so hard. 6K still leaves you some oversampling for a nice 4K delivery and I don't imagine you would need to punch in to 4K to stabilize a Ronin, it seems a bit extreme.

    As far as menus go: Menu > Overlays > Guides -- this is where you can set the frame guide to whatever size you want, either by aspect ratio and scale under "User" or by actual pixel resolution in "Absolute". A 6K frame would be 75% scale, a 4K frame would be 50% scale. Otherwise you can type in the exact pixel resolution for 6K or any other resolution you please. There's some other things you can play around with in this menu for your frame guides, like outline color and outside-of-frame shading and stuff.

    As always, though, these are just suggestions. If you can manage, I would run some tests for yourself and see how much stabilization you personally need to be happy and that'll inform how you will set your frame guides.
    Reply With Quote  
     

  5. #5  
    What Aaron said. Be careful you don't compromise all your footage just so you have the option to stabilise in post. Find a happy medium for the shutter speed so your footage isn't stuttery. I'd suggest adding frame guides at about 85% and framing within that.
    Reply With Quote  
     

  6. #6  
    A few things... You can to some extend if you got good post ad motion blur in post... you can never remove. So high frequency shake introduce motion blur to the shot... that will not look pretty when stabilised. Sure a short shutter brings the motion blur down But you still and especially with 8k cameras and especially with helium have rowing shutter issues. Those are really difficult to get rid of. Basically shaking your camera and stabilising it in post will add a pumping blurriness from the motion blur and also you will get a vertical gelly effect on the image. when vertical shaking and skewing when shaking on the horisontal axis....

    So no, your theory does not work. The only shakes you can counter steer this way is extremely smooth shakes. Like you can stabilise the horizon when on a steady cam etc. But the more hand held high frequency shakes you will not get around.


    best option out there now is to know your lens and also your rotations. That you can do with something like a SteadyXP... quite fiddley to set up but calibrated correctly for the given lens and camera you can basically neutralise shake, rolling shutter etc. But you loose on resolution and your motion blur will still be a smear...


    So short shutter and a steady XP mounted and calibrated for the lens and camera is best option possible... but its still a shity option. There is not for no reason steady cams and gimbals have been invented.
    Björn Benckert
    Creative Lead & Founder Syndicate Entertainment AB
    +46855524900 www.syndicate.se
    Flame / VFX / Motion capture / Monstro
    Reply With Quote  
     

  7. #7  
    Member Chad Peter's Avatar
    Join Date
    Mar 2007
    Location
    Los Angeles
    Posts
    82
    Hi Tom,

    We use a 15% pad on the 4th season of Mr Robot. On Mindhunter and House of Cards it was a bit greater than that, even - though I don't recall the exact percentage we used (18%?) My first question is - why 8k WS instead of 8k full frame? the WS option is simply cropping off more room to "stabilize".

    Not that this is helpful: but I'd use caution before fully adopting this workflow, because it's intensive and requires some training to pull it off properly. If you're goal is to simply use built-in stabilization plug-ins, you'll trade control & quality for speed. Sometimes those options work well, but it's not something we've ever considered on any of our shows. The proper workflow is to NOT stabilize within an NLE, but rather to export a DPX or EXR image sequence to work on in compositing software (like after effects), then we TRACK and REANIMATE the image, manually. We use Syntheyes for tracking (primarily), or Mocha (for more broad-strokes tracking), then export the 2D transform data into After Effects and re-build the X & Y coordinates in the After Effects graph editor.
    Reply With Quote  
     

  8. #8  
    Quote Originally Posted by Chad Peter View Post
    Hi Tom,

    We use a 15% pad on the 4th season of Mr Robot. On Mindhunter and House of Cards it was a bit greater than that, even - though I don't recall the exact percentage we used (18%?) My first question is - why 8k WS instead of 8k full frame? the WS option is simply cropping off more room to "stabilize".

    Not that this is helpful: but I'd use caution before fully adopting this workflow, because it's intensive and requires some training to pull it off properly. If you're goal is to simply use built-in stabilization plug-ins, you'll trade control & quality for speed. Sometimes those options work well, but it's not something we've ever considered on any of our shows. The proper workflow is to NOT stabilize within an NLE, but rather to export a DPX or EXR image sequence to work on in compositing software (like after effects), the we TRACK and REANIMATE the image, manually. We use Syntheyes for tracking (primarily), or Mocha (for more broad-strokes tracking), then export the 2D transform data into After Effects and re-build the X & Y coordinates in the After Effects graph editor.
    Thank you for this!
    Michael Tiemann, Chapel Hill NC

    "Dream so big you can share!"
    Reply With Quote  
     

Tags for this Thread

View Tag Cloud

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