Does this make sense? Or do I have just to copy the bitrate of the original file knowing that the quality haven't changed even if it was compressed? Last question. I can also do 50% just by dividing the original file to 0.5. This means I am allotting 35% more data to compensate for the missing data compressed by vp9. First I compute for supposed-to-be missing percentage of the h264 codec by getting the file size of the original file and dividing it by 0.65 (this means 35%) and applying the formula used in 2-pass option in ffmpeg documentation (I am using 2-pass option to accurately achieve computed file size). In my case, I use CBR by using " -x264-params "nal-hrd=cbr" ". So if I were to re-encode a vp9 clip, do I need to compensate the missing bitrate compressed by vp9 codec? I did try to use -qp option and it was unreasonable for me because if I use constant qp the bitrate goes so much higher and I think it's just a waste of storage without getting any more quality out of it and I also have no plan on using crf because I saw some articles that sometimes it ruins some part of the video. I know that every time you re-encode something the quality degrades but my question is do I have to compensate for the compressed data made by the vp9 codec to retain most quality? As far as I know vp9 codec is good for compressing video without loosing quality (about 20% to 50% efficiency). I am re-encoding my 2K videos from YouTube to H264 codec (YouTube only have vp9 codec for high resolutions) and I have a question to ask.
0 Comments
Leave a Reply. |