My processor, an Intel Celeron J3455, supports hardware decoding/encoding H.265 at 8-bit but only hardware decoding for 10-bit.
This is generally working but I've run into an issue when trying to convert an existing file from 10-bit H.265 to 8-bit H.265. Those were the bad ole days, but we loved them.I'm trying to convert my library from various formats into HEVC 8-bit mainly to shrink my library down. I used to use 2 pass when encoding DVD (mpeg2) episodes of Buffy and Lost Girl (Lot of night scenes, lot of kung-fu action, hard shadows) and it helped but took hours back then, and you had to fit the eps on CDROM. You can try it and see if it helps visually and seems 'smoother' in the action sequences, but it is a big time increase per file. It is better than a single pass that has a less localized compression rate when you want smaller files. 2-pass optimizes the compression level dynamically so there is less compression (and therefore distortion) in the highly kinetic and detailed scenes with sharp edge delineations. The single pass vs two-pass debate is not a thing for high-quality, high-bitrate encodes.It helps for high-compression encodes, and when file size is an issue.
You can try veryslow but it takes longer and looks barely better, Though highly kinetic action scenes exhibit "slightly" less blur I guess, but then action scenes move along so fast you really don't notice it unless the original is potato quality to begin with.-crf 18 and -crf 20 may be about the same, with the exception of very dark scenes. for one, a fast action sequence for another, a dark moving sequence or nightime scene, and a very finely-detailed scene with lots of things in it but not a lot of motion, perhaps people in a dining room talking or in an office across a desk.įfmpeg -i input.mp4 -c:v libx264 -profile:v high -preset slow -tune film -crf 18 output.mp4 A brightly colored video with blue sky, peoples faces, etc. I would suggest taking about 4 different sections of video with different visual profiles and seeing what works for each and all. Then experiment with parameters encoding into x264 at various settings until you are satisfied it looks "the same" as the x265 video you started with. So you make a trade-off in terms of time, file size, and computer power to achieve a nice re-encoding of the video stream. What you are trying for is "indistiguishable quality" from one encoding to another.
0 kommentar(er)
