Jump to content

Arri Amira 3.2K vs UHD


Mark Khalife

Recommended Posts

Hello,

 

I am shooting a short film on the Arri Amira. We will be finishing in 2K 1.85:1 (1998x1080) but I am considering filming at a higher resolution, either 3.2K or UHD (Pro Res 4444 XQ).

 

My question is, does anyone know if either of these 2 resolutions would downsample better than the other to 2K 1.85:1? This step would happen in Resolve.

 

It would probably be better for us to shoot at 3.2K for data management purposes but I want to make sure that we're not compromising anything.

 

Does anyone have any thoughts on the subject?

 

Thanks.

Link to comment
Share on other sites

Thanks Kenny for the reply.

 

Yes I am aware of that, but are you sure the downsampling to 2K would still be similar? Since neither of these larger resolutions is a multiple of the 2K 1.85 resolution, I'm just curious about the way they get downsampled, and whether one would downsample better than the other - even though the UHD is scaled up from 3.2K.

 

Thanks again.

Link to comment
Share on other sites

  • Premium Member

If 3.2K or UHD both come from the same 3.2K sample, then why would the version that upscales to 3.8K before going back down to 2K be better than going directly from 3.2K to 2K? And if it were better (unlikely), then what's to stop you from upscaling it to 3.8K yourself in post before downscaling to 2K? I mean, it's possible that the recorded UHD version is better to work from because the upscaling happens before compression to ProRes, but it may not be an improvement that is particularly noticeable. I guess it's worth testing, but my gut feeling is that one downscale step is better then two steps (one up and then one down) to get to 2K.

Link to comment
Share on other sites

  • Premium Member

I can't say I've noticed any real advantage to 3.2k Prores over the internally downscaled (and genuinely 4:4:4) 2k Prores for a 2k finish.

 

It's great if you're worried about reframing, or for adding post-zooms. But you're not getting same level of per-pixel quality with 3.2k straight off the sensor (which is effectively offering you 4:2:2), versus the downscaled and properly subsampled 2k 4:4:4.

Link to comment
Share on other sites

  • Premium Member

I wouldn't think of ProRes 3.2K 4:4:4 as actually being 4:2:2 -- you'd just expect some difference in practical resolution between the three channels compared to 2K 4:4:4 where there would be similar resolution in all three but the luminance signal just being lower in resolution than it is in 3.2K. I would think that in 3.2K 4:4:4 is true 4:4:4 in terms of bandwidth or else they wouldn't call it 4:4:4.

 

Besides, if you end up downscaling in post from 3.2K for a 2K finish, rather than in the camera, then what's the disadvantage of recording 3.2K other than the increase in data? Either way you will reap the same advantages of getting to 2K 4:4:4 other than the compression issues of going ProRes rather than directly from the camera's sensor internally.

 

But as you said, unless you (A) might need a UHD master, or (2) want to do resizing and/or stabilization to shots, then you might as well shoot 2K if you are finishing in 2K. If I were doing an action film where stabilization was more common, or something with a lot of visual effects but raw recording wasn't an option, then 3.2K might make more sense.

Link to comment
Share on other sites

Mark,

 

I think that the only way to make this decision is to perform a test of each workflow.

 

You'll want to test two scenarios:

 

1. Record 2k vs. record UHD with no cropping or repositions

2. with repositions and with zooming in as this sometimes occurs.

 

then see if you can see any practical difference. I suspect that it might be difficult to see any meaningful differences, but who knows?

 

Also, keep in mind that when shooting at 24fps maybe one out of 36 frames will be free of significant motion blur. So very few frames actually have 2k, 3.2k or even 4k resolution. Of course, static shots without movement will show the maximum resolution possible.

 

So, when you test, don't just shoot static test frames. Also shoot something with pans, camera moves and subject movement to really get an idea of the differences in the capture workflows.

 

In reality, if you don't have access to the camera and post facilities to conduct the test, you'll always have a sense of doubt about choosing to record 2k, and I suspect that you'll choose to capture in UHD, with the additional expense of doubling the hard drives you'll need for capture and back-ups. And, perhaps, you'll also need the additional step of transcoding everything to 2k or HD for editorial as well.

Link to comment
Share on other sites

Thank you everyone for the responses, very much appreciated.

 

My main reason for shooting 3.2K or UHD is the (slightly) larger sensor area used and the ensuing larger angle of view. Also, through my experience with originating in 4K (on different cameras) and downscaling to 2K, I've noticed a slightly sharper image than a 2K origination, which I normally like.

 

As David said, I was indeed wondering about the scaling pre-ProRes, and whether it mattered. But tests are definitely in order.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...