Isn’t VP9 more efficient, yielding better quality at a smaller size? Yet it still seems most people don’t use VP9. If this is the case, why? I know youtube re-encodes in VP9.
Does it matter if the content is 1080p vs 4k in terms of which codec you would choose to export with?
Thanks. I didn’t know you couldn’t export VP9 as an mp4 file. If I’m filming with a smartphone, the phone is likely compressing it with H264. Is it silly to export a project using that footage in ProRes or something better than the original? Or will that prevent double compression when I export in shotcut with H264?
Well, you can’t export better than the original source. You either use a lossless codec or use the same as the source with the same compression ratio. You won’t get ‘double compression’, just re compression.
So if I export with the same codec and the same bit rate as my source material, I shouldn’t have quality loss or added artifacts? I assumed every time you decompress and re-compress something you would lose some quality in the process due to entropy. Maybe I have that wrong.
“You won’t get ‘double compression’, just re compression.”. Is it likely that the re compression will cause quality loss? If I were to export with the same bit rate and same H264 codec as the original source.
Yes. But barely noticeable if your source video is of high quality.
When you unpack (de-code) then re-pack (en-code) you will always get some wrinkles. The more often you do it, the more wrinkles and the easier it becomes to see them. Do it once and you’ll barely notice (given that you don’t try to en-code too tightly to get a tiny export!)
I also always wondered why the link that Steve provided suggests H264, and not VP9. But, when I upload to YouTube, the video is not rendered to VP9 immediately; it is rendered to a video and audio codec other than VP9/AAC (don’t recall which at the moment), then after a variable period of time, the video is converted to VP9/AAC.