Overview: When measuring video quality with metrics like VMAF you should measure and report consistency (transient quality and standard deviation) as well as mean quality. If you’re interested in the quality consistency of AV1 codecs and typical configuration options, you can attend a webinar ($149) or take a course on AV1 production that I will…
Jan Ozer here. I spent the last 2 months working with four AV1 codecs, benchmarking them against each other and x264 and x265. I’m presenting my findings in this for-fee 75-minute webinar ($149.99). There are two parts; the first shares what I learned about the codecs, and the second details what I learned about encoding…
Setting your I-frame interval in seconds rather than frames allows you to use one encoding string for files with varying frame rates, saving time and minimizing the risk of error. Author’s note: See a new section regarding a LinkedIn comment that this technique “interferes with the rate control mechanism of codecs.” I just finished a…
While standards-based video codecs like H.264 and HEVC tend to dominate, standards-based video quality metrics have never risen to the same usage or attention level. With two innovative and highly accurate metrics now available from the ITU-T, this may change in the near term. Briefly, these models are: ITU-T Rec. P.1203, which calculates the quality…
Just a quick note to let you know that I’ve published a comprehensive report that documents the testing, quality, and performance of the Low Complexity Enhanced Video Codec (LCEVC) as developed by V-Nova and in the process of becoming standardized by MPEG. By way of background, I’ve been consulting with V-Nova to test LCEVC using…
Previous tests presented here showed that a Windows version of FFmpeg compiled by Zeranoe was about four times slower than a version compiled by a colleague who works in an OTT organization. Since then, Miguel Perez contacted the FFmpeg organization and Zeranoe, who researched the issue, recompiled, and created a new download that resolves the…
AV1 encoding on Windows is now only 2x slower than x265. With a properly compiled version of FFmpeg, encoding performance is slightly faster than the Alliance for Open Media’s encoder, though output quality is very slightly lower. Congrats to AOM for delivering on their promise to get encoding times down. Thanks for all the help…
TL/DR: Netflix updates VMAF to include a new model that reverses out the effects of techniques like sharpening, contrasting, and histogram equalization. As I reported in VMAF is Hackable: What Now?, you can increase VMAF scores via simple image enhancement techniques like sharpening, contrasting, and histogram equalization, which, when applied inappropriately, may not actually improve…
AV1 encoding is now only 2x slower than x265. See the latest results here. I began testing AV1 early this week. Briefly, my tests involve 16 ten-second clips in four genres (movies, sports, animations, gaming) and an “other” category (music video, nature video). I’ve completed the first set of tests with FFmpeg 4.3, benchmarking x264,…
I’m reviewing AV1 encoders for Streaming Media Magazine and was performing some QC on the AV1 files that I produced with the latest version of FFmpeg. Two file details I always check are GOP size and whether the GOPs are closed or not. I check these for multiple reasons. First, FFmpeg doesn’t have the most…
