Gianni Rosato

@giannirosato.com

150 Followers 178 Following

Multimedia Compression Engineer Mastodon: @gianni@disobey.net Web: https://giannirosato.com/ Work: https://halide.cx/

Iris-WebP encoder performance on a BD-rate to speed graph. Encoders closer to the top left of the graph are better. Iris-WebP is at the Pareto front for image compression in SSIMULACRA2, a perceptual image metric with high correlation with subjective quality.
“[sorry for the spam] Just wanted to say that theguardian.com is on `tune=iq` for the last two weeks (we serve AVIF to non-JXL-supporting browsers) and we couldn’t be more happy. Encoder consistency is much, much higher (still not as good as JXL, but that’s just an impression, not a rigorous assessment): some highly compressible images where delicate detail was being obliterated grown even ~3× in filesize and look much better while unnecessarily weighty images shrank, mostly without any perceptible degradation. Interestingly, we kept the same quality values (which differ by dpr) and as much as we can say, the overall bandwidth stayed more or less the same (I can’t see any change in the graph), so it’s likely the massive reduction of rare weighty images offset small images growing. Huge thanks to juliobbv for his work on `tune=iq` and to whomever else worked on it. And to friends at our image CDN (Fastly). And to everyone here as without this server, I wouldn’t even know about it… As always, we welcome improvements to any and all encoders. So the idea of “splitting Julio” couldn’t sound more exciting 😉.”
Post: “This is one of the craziest ideas l've ever seen. He converted a drawing of a bird into a spectrogram (PNG -> Soundwave) then played it to a Starling who sung it back reproducing the PNG.” Response: “Ok fine. Sure. Whatever. We can save PNGs to birds. I give up. You win world”
Iris-WebP BD-rate performance, showing speed-to-quality efficiency roughly on par with standard libaom.
"A WEBP file is an image saved in the WebP (pronounced "Weppy") raster image format developed by Google for ..."
A meme about the XPSNR video metric, making fun of it
Eve-VP9 versus open-source alternatives (libvpx, x264, x265) showing impressive gains.
STOP DOING CODECS • VIDEOS WERE NOT SUPPOSED TO BE GIVEN LETTERS AND NUMBERS • YEARS OF ENCODING yet NO REAL-WORLD USE FOUND for ANYTHING MORE THAN H264 • Wanted to record videos in higher quality? We had a tool for that: It was called A REEL OF FILM • "Yes please give me 10 bit 4:2:2 chroma. Please give me Rec. 2020 color" - Statements dreamed up by evil wizards LOOK at what encoding nerds have been demanding your Respect for all this time, with all the h264 compatible media players we built for them (This is REAL encoding Y' Cb+Cr Y'CbCr science done by REAL encoders ): Duration: 00:00:05.13, start: 0.000000, bitrate: 389334 kb/s Strean #0:0[0x1] (eng): Video: prores (4444) (apth / 0x68347961), yuva144p12le + 4:4:4 - N5000* RG8 /s, SAR 1:1 DAR 256:135, 30 fps, 30 tbr, 30 tbn (default) creation_time : 2024-10-27123:05:84.000000Z + 4:2:2 -ClOaK handler_nane : Apple Video Media Handler encoder : appl : Apple ProRes 4444 + 4:2:0 Strean #0:1[0x2](eng): Data: none (tacd / 0x64636074), ® kb/s (default) Metadata: creation time : 2024-10-27723:05:04.000090z handler_nane : Tine Code Media Handler 4:1:1 :00:00:00:00 insupported codec with id 0 for input strean 1 ????? ??????? ????????????????? "Hello I would likeDNxHR apples please" They have played us for absolute fools
The real reason the US government doesn't want you to use C. And you thought it was because of "memory-safety"? [shows various languages and their corporate ‘stakeholders’ alongside ‘uncontrollable’ C]