I recently made a video to commemorate my daughter’s 16th birthday, and I wanted the video to be of the highest possible quality, so I decided to use 4K (technically UHD, 3840x2160) resolution. The videos from my daughter’s first few years were recorded in 720p or 1080p, so I researched methods to upscale videos for the purpose of getting higher quality videos from Final Cut Pro (FCP). Bottom line: while FCP can upscale videos, you’re better off with a third party video upscaling application like Topaz Video AI. What I also found out was that all these years I’ve been exporting videos from FCP the wrong way.
Normally I’d export my FCP projects using the H.264 codec. I learned that it’s better to export from FCP using one of the Apple ProRes codecs (Apple ProRes 422 offers a good balance between quality and file size) and then use Compressor to convert the ProRes video to the H.264 codec. FCP uses Apple’s internal H.264 encoding engine which is optimized for speed, but it doesn’t always use the most efficient compression passes or bit allocation strategies. It may use a single-pass variable bitrate encode with conservative parameters which results in a quick export but can produce larger file sizes and may be associated with some quality loss. On the other hand, exporting in Apple ProRes is an intra-frame codec that yields a high-quality master, and feeding that master to Compressor has quality advantages. Using the Social Platforms > 4K preset, Compressor uses multi-pass encoding and performs better motion analysis and bitrate distribution, giving more bits to complex scenes and fewer to static ones. Here are the file sizes for each workflow:
1-STEP WORKFLOWFCP export with H.264 codec: 8.44 GB
…followed by Compressor conversion to H.264 codec: 5.15 GB
In full screen mode on a 32-inch UHD 3840x2160 monitor, I cannot see any difference in quality between the two H.264 files. In reality, there are some technical advantages with the 2-step workflow. FCP’s built-in exporter prioritizes speed over compression precision, which can lead to slight quality loss. The 2-step workflow is superior because Compressor encodes from a pristine, uncompressed ProRes master using more advanced multi-pass analysis and smarter bitrate allocation, producing a cleaner, more efficient H.264 file. It’s like the difference between working with an original photo vs. a photocopy. Perhaps a more discerning set of eyes could tell the difference in quality, but I’ll just conclude that the 2-step workflow yields a file whose visual quality is at least as good as or possibly better than the 1-step workflow.
So in summary, the 1-step export from FCP to H.264 is faster but comes at the expense of larger file sizes and possibly some slight degradation in quality, whereas the 2-step process involving FCP export to ProRes, followed by Compressor conversion to H.264 takes longer but results in smaller file sizes and equally good if not slightly better quality. I don’t mind waiting a few minutes for Compressor to convert videos, so unless I’m feeling particularly impatient, I plan to use the 2-step workflow from now on.
For more on this topic, see this article. Happy video encoding!
