Announcement

Collapse
No announcement yet.

VVenC 1.11 Brings More Performance Improvements For H.266/VVC Encoding

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Originally posted by avis View Post

    I was 200% sure someone would post this asinine evil stupid meaningless comment. Kudos. I'm not disappointed!




    None. From 38 to 46% bitrate savings over AV1. Absolutely nothing.

    A HW decoder with a lot less complexity requiring probably half the AV1 transistor budget. A lot more power efficient decoding as a result.
    Oh, that seems pretty significant. Would you care to cite your source?

    Comment


    • #32
      Originally posted by avis View Post

      None. From 38 to 46% bitrate savings over AV1. Absolutely nothing.
      I prefer a different way of comparing quality between different encodes.

      Since PSNR literally stands for Peak Signal to Noise Ratio and since traditionally only the Y channel of I frames is looked at, the results tend not to match subjective quality meaurement.

      The same holds true for SSIM, MS-SSIM, VMAF, pretty much any metric.

      What i prefer to do is use MSU's Quality Measurement Tool, I use the Windows version because it has an easy GUI, and i calculate PSNR/SSIM/MS-SSIM/VMAF across all frame types, I/P/B and all channels, YUV, and i dump the results in an Excel spreadsheet.

      I then run a statistical analysis on each to see how many frames where above a certain metric, say 45dB for PSNR, what the min and max values were, what the average was and so on.

      This gives a much more accurate view of how quality compares.
      Last edited by sophisticles; 21 February 2024, 02:46 PM.

      Comment


      • #33
        Originally posted by avis View Post
        Um, a rare codec aficionado. The problem with you is that AFAIK you're also deep in the "We hate" camp. Not a single comment or thread in regard to VVC And of course none of your comments are phoronix. Videohelp is an entirely different forum where people don't swear allegiance to codecs based on their patent purity.
        Videohelp and Doom9/Doom10 didn't have people that swore allegiance to codecs based on patent purify but they swore allegiance to one encoder, x264, based on the cult of personality known a Dark Shikari, aka Jason aka Fiona.

        He/she commanded the same type of blind allegiance as Linus does here, if he said it then it was gospel, and if you disagreed with it it was because you were stupid, misinformed, poorly educated and/or troll.

        Comment


        • #34
          I still don't understand. Why do even people bother to check the comments section for this news piece and why do they need to express their desire to use a totally different unrelated codec? Why do you wanna get heard so much? There are mostly anonymous random Joes here who couldn't care less about your AV1/AV2/whatever codec preferences. We all know that anything patent encumbered is "garbage" even if it has billions of users or uses.

          Originally posted by jonkoops View Post

          Oh, that seems pretty significant. Would you care to cite your source?
          In recent years, an interest in multimedia services has grown expeditiously where the main part is comprised of video. Firms as well as subscribers require higher resolutions, framerates and sampling precision which results to higher amount of data need for processing, storing and transmitting. Therefore, a big challenge arises for researchers to develop new compression standards which should reduce the huge amount of data and keep quality at the same level. This paper examines compression performance of the latest and currently most used video codecs, namely H.266/VVC, AV1, H265/HEVC and H.264/AVC. The test set consists of seven sequences with various content at 8K, Ultra HD and Full HD resolution and encoded to bitrates from 1 to 15 Mbps for Full HD and Ultra HD resolutions and from 5 to 50 Mbps for 8K resolution, respectively. Codec performance was measured using PSNR, SSIM and VMAF objective quality metrics. In terms of the Bjøntegaard-Delta (BD) model, the results showed that H.266/VVC outperforms all other codecs, namely H.264/AVC,H.265/HEVC and AV1, respectively. Averaged bitrate savings were approximately 78\% for H.266/VVC, 63\% for AV1 and 53\% for H.265/HEVC relative to H.264/AVC, 59\% for H.266/VVC, 22\% for AV1 compared to H.264/AVC and 46\% for H.266/VVC relative to AV1, all for 8K resolution. The results also showed that the performance varies depending on resolution – with higher resolution, the efficiency of newly developed codecs such as H.266/VVC and AV1 is greater, which confirmed the fact that the H.266/VVC and AV1 codecs have been primarily developed for videos at high resolutions as 8K and/or UHD.
          Last edited by avis; 21 February 2024, 03:02 PM.

          Comment


          • #35
            Originally posted by avis View Post
            I still don't understand. Why do even people bother to check the comments section for this news piece and why do they need to express their desire to use a totally different unrelated codec? Why do you wanna get heard so much? There are mostly anonymous random Joes here who couldn't care less about your AV1/AV2/whatever codec preferences. We all know that anything patent encumbered is "garbage" even if it has billions of users or uses.



            https://www.preprints.org/manuscript/202402.0869
            We just want to promote enthusiasm for free codecs like AV1 and pooh-pooh licensed codecs like H.266 in hopes that the sentiment will reach people in hardware manufacturing companies before too many start using it. For instance, I would love it if my security cameras used AV1 rather than their current H.265. And I hope that the newer products start to use AV1 rather than H.266.

            Comment


            • #36
              Av2 grinds vvc in to the dust …

              CodecWar is an analysis service created by ViCueSoft, the developer of codec analy­sis tools VQ Analyzer, VQ DVK, and VQ Probe. As currently configured, the site's ideal users are researchers who are looking for a convenient way to compare codecs using relevant datasets and codec developers who are looking for a structured way to benchmark their codecs against others. In contrast, it's not a particularly convenient way for streaming producers to run experiments to optimise their encoding parameters, although it could grow into this.


              …not that it matters - we’re still waiting for a use case that h264 isn’t good enough for…

              Comment


              • #37
                Originally posted by OneTimeShot View Post
                Av2 grinds vvc in to the dust …

                CodecWar is an analysis service created by ViCueSoft, the developer of codec analy­sis tools VQ Analyzer, VQ DVK, and VQ Probe. As currently configured, the site's ideal users are researchers who are looking for a convenient way to compare codecs using relevant datasets and codec developers who are looking for a structured way to benchmark their codecs against others. In contrast, it's not a particularly convenient way for streaming producers to run experiments to optimise their encoding parameters, although it could grow into this.


                …not that it matters - we’re still waiting for a use case that h264 isn’t good enough for…
                AV1 is already more computationally expensive than VVC despite offering a worse compression ratio.

                AV2 will be even more computationally expensive (normally new codecs are an order of magnitude more expensive to encode and at least twice expensive to decode) and probably outside the reach of most people. Let's celebrate it! And someone will have to pay for that regardless.

                Comment


                • #38
                  Originally posted by Chugworth View Post
                  We just want to promote enthusiasm for free codecs like AV1 and pooh-pooh licensed codecs like H.266 in hopes that the sentiment will reach people in hardware manufacturing companies before too many start using it. For instance, I would love it if my security cameras used AV1 rather than their current H.265. And I hope that the newer products start to use AV1 rather than H.266.
                  There are no people related to making such decisions here. This is an echo chamber of people who believe Linux/Open Source/Patent Free world is the best there is.

                  Except people who create closed source software or invent patented stuff are those who want to make the world a better place. And you are just a consumer.

                  And let me reassure you. Despite one or two AMD developers lurking there, whatever people say here will not affect AMD decisions in terms of developing new HW. This is not going to happen not matter how much you spam in VVC related topics. You don't even know how much AV1 IP core costs and it's not free. Do you even know what IP core is? No? Darn, how embarrassing.

                  Not a single word in the topic about VVC encoding/decoding/compression efficiency or performance. Over 35 off-topic messages, half of them praising AV1 and reviling VVC. Absolute cringe. I could have understood the sentiment had we have teenagers here but I'm certain the vast majority of people here are at least 25 yo. Blindingly hating anything made by MPEG. Anything patented. No one even remembers that it's MPEG stuff which made home video/audio available for billions of people (MPEG2 video/MPEG2 Layer 3 aka MP3 audio). No? Don't remember? Let's sweep history under the rug?

                  And of course no one remembers that AV1 uses some algorithms which were first patented and then patents expired. Nah, AV1 is "clean" and created by God and doesn't use anything that once was part of MPEG.
                  Last edited by avis; 21 February 2024, 03:45 PM.

                  Comment


                  • #39
                    Originally posted by avis View Post

                    AV2 will be even more computationally expensive (normally new codecs are an order of magnitude more expensive to encode and at least twice expensive to decode) and probably outside the reach of most people..
                    Do you have a source to back that conjecture up? It isn’t a surprise that VVC is basically AV1.5 because only Google do any R&D.

                    Comment


                    • #40
                      Originally posted by OneTimeShot View Post

                      Do you have a source to back that conjecture up? It isn’t a surprise that VVC is basically AV1.5 because only Google do any R&D.
                      1. This is common knowledge among video codecs experts.
                      2. You can literally take MPEG2/H.264/H.265 reference encoders and compare their encoding speeds using default presets. Or VP8/VP9/AV1 if you so choose.

                      Sorry I didn't understand your second sentence or the irony therein.

                      But no, "only Google do any R&D" is a load or pristine BS. VVC has literally several hundred encoding tools (algorithms) - all very ingenious and smart, otherwise no one would have released it as a new codec.

                      Comment

                      Working...
                      X