REMUX VS HVEC 10BIT FULL

All encodings implemented 2-pass encoding with a set target bitrate. All sequences are 10 seconds long and used in YUV 4:2:0 subsampling.įor encoding, I used default settings with ffmpeg. Some of these sequences are well known and were already used in several prior standardization activities.

REMUX VS HVEC 10BIT FULL

Test setįor the test set I used Full HD and 4K sequences from the JEVT SDR test set which was also used in the standardization of VVC. If you want to recreate the same execution environment: I used Docker to build it so you can recreate the exact same environment using my Dockerfile which can be found here. I also compiled libvmaf (version 1.5.1) and ffmpeg (version 4.2.3) to run the encoders and perform PSNR, SSIM and VMAF measurements. x265 encoder (version 3.2) for h.265/HEVC encoding.libvpx-vp9 encoder (version 1.8.2) for VP9 encoding.This article delves into that experiment and shares the results of Christian’s research. These results were obtained using the open-source encoders libvpx-vp9, x264, and x265. ĭuring the session, we discussed the fundamental differences between the two “modern codecs” and tied it off with an early analysis of each codec’s performance.

It's not that you won't see it at all in x265, it's that it will be much worse in x264, even unacceptably so, where x2165 might be perfectly usable at the same bitrate.For an APAC live event, our video coding engineer Christian Feldmann compared the HEVC (H.265) vs VP9. At some point, x264 should have much more visible PQ loss then x265. Take your x264 bitrate, and reduce it by say, 30%, and encode with x265 at the same rate and see what what it looks like then. So the way that you actually test this is in fact the flip of what you were doing. If you have the same source, and sufficient bitrate for different compression types, the older one will probably look better, because it's not discarding as much. But of course we don't do that because lossless would be gigantic. The easy way to remember this for me: what is the absolute best picture quality you will get? The logical end point of the example is: lossless. In fact, if you are after better picture quality, mpeg-2 would be better looking then h.264, but the catch is only _if you have sufficient bitrate_. But at the end of the day, you are still discarding information. Newer compressions (h.265 vs h.264) have more tricks to mask the this loss. Consider that modern compressions of this sort are lossy schemes, that is, they work by discarding more and more of the image with the goal that you won't be able to notice. In fact, and this is somewhat non-intuitive at first, if you have sufficient bitrate, generally older should be better as far as picture quality is concerned. Newer is generally better at not showing picture quality degradation as you constrain the bitrate, i.e force smaller files. The way higher and higher compression works is more nuanced then 'newer is better'. Second, your understanding is based in an incorrect and incomplete understanding of the compression. Do you have a 10-bit capable display? Because if you don't, 10-bit potentially will look worse. You also have a lot of factors when you start talking 8-bit vs 10-bit.

First up, for your encodes, did you have 10-bit sources? If you didn't have 10bit sources, there's no point doing a 10bit encode, you'd have 2 extra bits allocated for precision you don't have.

Remux vs hvec 10bit