-
lossless-compressed footage looks bad when scaled up
I recorded a screenshot video (using ScreenFlow) as a quicktime .mov file, with the lossless “Apple Animation” codec in the .mov file. The footage shows a browser as a URL is typed and a webpage loads. It looks good when I open it in quicktime, pixel-perfect.
I’ve imported the clip in Final Cut 5.1, and I’ve zoomed it in (the original footage is the full-screen, but i want to use a small frame centered on the browser URL), using the Scale slider on the motion tab. The quality looks really bad– i.e. if I zoom-in with Photoshop on a freeze-frame from this clip, the quality looks great (becase photoshop uses a high-quality scaling algorithm with bicubic sampling, and the raw footage is pixel-perfect with no compression artifacts), but FCP’s basic scaling algorithm on the motion tab seems to be really low-quality. Is there a better way to scale this in FCP? Should I use Apple’s Motion program? Or perhaps a 3rd party plugin in FCP?
The reason I can’t just use photoshop is that it’s not just a static shot, there’s movement.
-G
-
3 Replies
-
Geoffrey,
You haven’t mentioned what your timeline settings are…
If you’re final is SD, you should do your screen capture at HD resolution, thus giving yourself a source several times larger than SD that you can zoom in on without pixelation. And, hopefully you’re not trying this on a DV timeline, which will also create havoc with any graphic…
David
David Roth Weiss
Director/Editor
David Weiss Productions, Inc.
Los AngelesPOST-PRODUCTION WITHOUT THE USUAL INSANITY ™
A forum host of Creative COW’s Apple Final Cut Pro, Business & Marketing, and Indie Film & Documentary forums.
-
It shouldn’t make a difference, but they’re both the same HD resolution.
One more clarification: I’m not concerned with lower quality in my RT quick renders. This is happening in my final render. However, there must be a setting somewhere that is lowering the quality.
Whenever you zoom-in a raster graphic, there are several algorithms to do it (see https://en.wikipedia.org/wiki/Bicubic_interpolation). Photoshop seems to be using bicubic, whereas Final Cut seems to be using bilinear.
Is there a different setting, or effect, or plugin to get better results from Final Cut?
I’m going to go re-check every single quality setting that could be affecting the clip.
-
update: i think i understand the problem now, and i’m hoping somebody knows the solution.
The project is HDV1080i. Most of the footage in the project has that setting. But this screen-capture clip is NOT interlaced. So I believe the “Basic Motion” scale operation is expecting interlaced footage.
I tried changing the field dominance settings of the project and the clip, in every combination, and the artifacts changed, but nothing that looked better. I applied a FlickerFilter-Max, and it looks a lot better. But still not good. I wish there was a FlickerFilter that could go past max.
Anybody know about mixing progressive footage into an interlaced project?
g
Reply to this Discussion! Login or Sign Up