Question on resolution and editing in FCPX

I know this is probably a very elementary question, but I am still in grade school when it comes to video. If I shoot in a higher resolution and then edit in a lower resolution, how can I calculate how far I can enlarge the image without degradation? Effectively, I am zooming in by using the transform part of FCPX

If I am editing in 720p and have video in 1080, 2.7K or 4K do I just divide the 720 into the higher resolution to get the maximum percentage of "zoom"? For example, 1080/720 = 150%

Thanks!

My personal rule of thumb, never enlarge over 120%, or you will start to see pixelation.

Professionally, I'd determine my delivery format first. If it is 1080, everything gets shot in 1080 minimum, and I edit in a 1080 format timeline. All of this should be planned out before you shoot.

But, we don't live in a perfect world. I'd never enlarge 720 to anything over 1080. I'd never enlarge 1080 to any "K" resolution. Once you get too big, pixelation will kill your image quality.

in addendum to Ben's excellent rule-of-thumb ..


I do like re-framing a lot = me bad camera-man

trouble with video is: it's not like 'pictures', with its dots, each containing all info needed.

Trouble starts in your cam: most consumer devices use so called Bayer-filter . Your cam has twice as much green than red & blue! (Which is , btw, the reason for green-screen, instead of, from the analogue days, blue-screen)

Next, compression: most consumer devices use a chroma-subsampling of 4:2:0 - in my own ignorant words:
The color-resolution, compared to the luma resolution, of your cam is only half (a quarter?!) than 1080!

So, when you zoom into your 1920x1080 video, you don't have 1920x1080 pixels to enlarge.... :woohoo:

but - that's all theory. It depends too on content and 'need'... when you have to zoom-in by 500% to proof Big Foot in your recordings of your latest trip thru the woods - just do it!

Dividing 1080 by 720 would be a good way to get the scaling limit. In theory that 150% would get you back to 1:1 1080p. I'm not sure in the real world if that leads to any quality difference vs setting Spatial Conform to "None" and scaling the 1080p footage down. I've been happy enough with scaling up that I've never bothered to do a comparison.

As others have said, if you're scaling lower resolution footage up, then 120-130% is about the limit. having said that, I've been happy with footage scaled by as much as 160% if I get Resolve to do it.

"(Which is , btw, the reason for green-screen, instead of, from the analogue days, blue-screen)"

Actually not. Blue came first, then green, both before digital imaging. Both are as far from flesh tones as you can get, which is the reason for their specific formulations. Green is used more in digital, because the first digital image sensors were more sensitive to green, that part is correct. But Chorma Green was not developed "for" digital.

And Bayer-ing is used in high end cameras, also.

As for scaling and lower resolution vs higher resolution clips, well, that got me thinking. How often do you need to up-scale 4K images?

BenB wrote: ... But Chorma Green was not developed "for" digital....

I know that, one of the first chroma-key in commercial use, afaik, was Disney's 'Sodium Vapor Keying' (Mary Poppins...) - long before electronic color-keying.

But, working in TV industry (not on the technical side) in the 80ies, I was under the impression, they used Blue cause of its 'distance' to flesh-tones and therefor 'easier' to 'filter' from the analogue signals.. And, when chip-sensors with Bayer-filters appeared, due to its higher sensitivity and resolution, they switched to Green.

... obviously, a wrong impression. :whistle:

Yes, blue was developed/used first, and yes, green was more widely adopted recently because it works with digital sensors better than blue. But it wasn't created "because" of digital sensors. But before digital sensors, blue was much, much more used because on film it was furthest from flesh tones. So yes, pretty much you got it.