-
Text Pixelate with Projection
Hi there
I work with a church where they do regular video projection to screen.
The videos are normally edited either in XDCAM or ProRes 422 in FCP, with text added in AE and then rendered out as animation, after which the final video and audio will be rendered out of FCP in one of the above codecs.
After that, it gets encoded into a MPEG-PS (MPEG-2) file with a bit rate of 8000 (in Episode Encoder), to be played out from the WATCHOUT system that they use on Sundays. The projection is natively 1024×768, but the encoding also letterboxes the video so that it fits nicely onto the 16×9 screen as 1024×576.
The problem they’ve noticed was that very often the video will look fine, but the text thrown over the video will be quite pixelated.
We’ve wondered about the fact that sometimes, out of bad habit from the Z1 days, the editors still work in anamorphic and whether this could have an impact on the text.
Other than that, we’ve been a bit baffled as to why only the text would look jagged and not the video.Any help would be greatly appreciated.
Cheers
Mario
-
6 Replies
-
[Mario Sahe-Lacheante] “The videos are normally edited either in XDCAM or ProRes 422 in FCP, with text added in AE and then rendered out as animation, after which the final video and audio will be rendered out of FCP in one of the above codecs”
What comp settings are your AE artists using?
Walter Soyka
Principal & Designer at Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog – What I’m thinking when my workstation’s thinking
Creative Cow Forum Host: Live & Stage Events -
Hi
It varies, there’s no particular template everyone works off – because different people contribute to the media from different areas, including some freelancers, I know this much:
Most of them would render natively out of FCP and import into AE, mostly Apple ProRess 422 HQ and XDCAM from EX3 cameras, and work from then onwards to add effects and text.
So Comp Settings would be the native size of the video worked with in FCP, I think at square pixels for whatever amount of time at 25fps.
I have seen though, sometimes people will work in anamorphic in FCP, 16×9 an AE, and render out the final video again from FCP in anamorphic, before our Episode process makes it 16×9 again.
It’s really stupid way of working which I think comes from habit with years of working with Z1 footage, but I’m trying to get them to stop that. Again, this isn’t the norm, it’s about 50/50 at the moment with all our media.Short of that I don’t know of anything else that can have an effect.
Our problem we’ve noticed though is not AE as a whole. With effects and filters and stuff added, the only thing we notice that pixelate is text.
Thanks for your help and input!
-
[Mario Sahe-Lacheante] “XDCAM from EX3 cameras”
XDCAM EX SP (1440×1080), or XDCAM EX HQ (1920×1080)? Hopefully this is on the same setting for all cameras…
[Mario Sahe-Lacheante] “It varies, there’s no particular template everyone works off – because different people contribute to the media from different areas, including some freelancers”
I’d think that this is your first problem in trying to track down the issue. A standardized workflow will make QC issues like this much, much easier to find and to fix.
[Mario Sahe-Lacheante] “I have seen though, sometimes people will work in anamorphic in FCP, 16×9 an AE, and render out the final video again from FCP in anamorphic, before our Episode process makes it 16×9 again”
Yikes. There’s nothing wrong with working anamorphically all the way through, but bouncing back and forth will not be kind to your footage.
[Mario Sahe-Lacheante] “Most of them would render natively out of FCP and import into AE, mostly Apple ProRess 422 HQ and XDCAM from EX3 cameras, and work from then onwards to add effects and text.”
As a general rule of thumb, After Effects performs better with intraframe-compressed footage like ProRes than it does with interframe-compressed footage like XDCAM (which uses long-GOP MPEG2 for video). Also, ProRes uses 4:2:2 chroma subsampling, but XDCAM is only 4:2:0, so ProRes is much better suited for graphics.
I’d try to standardize your FCP editors on square-pixel ProRes (you probably don’t need ProRes HQ); that should smooth out your workflow down through the rest of the tool chain.
Walter Soyka
Principal & Designer at Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog – What I’m thinking when my workstation’s thinking
Creative Cow Forum Host: Live & Stage Events -
Thank you Walter
I will try to work down the issues you mentioned to find a uniform workflow for all of us.
The EX 3 footage is all HQ at 720p/ 1080p, we never use SP, but you made some points that I will work to integrate into what our guys do.
Thank you so much for your time.
Mario
-
I hope I was able to help. If you need some more input on AE-FCP workflow, there are lotsof knowledgeable people in the After Effects forum and the Final Cut forum here on the COW.
Walter Soyka
Principal & Designer at Keen Live
Motion Graphics, Widescreen Events, Presentation Design, and Consulting
RenderBreak Blog – What I’m thinking when my workstation’s thinking
Creative Cow Forum Host: Live & Stage Events -
The real way to do this is to use Photoshop to create PSD or PNG files with transparency and do the Text part in Watchout. Just load the text as a file in Watchout, add Opacity tween points to the text layer, and just fade it in and out as needed. Don’t worry about all the encoding settings as much. The good part about this system is that you don’t have to re-render all the footage every time you want to make a change to the text. You just open the file in Photoshop and edit the text.
Daniel Brodie
brodiegraphics.com
Reply to this Discussion! Login or Sign Up