{"id":182645,"date":"2026-07-15T16:53:50","date_gmt":"2026-07-15T14:53:50","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/vypusk-svt-av1-4-2-0-kodirovshhika-dlya-formata-video-av1"},"modified":"2026-07-15T16:53:50","modified_gmt":"2026-07-15T14:53:50","slug":"vypusk-svt-av1-4-2-0-kodirovshhika-dlya-formata-video-av1","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/novosti-interneta\/vypusk-svt-av1-4-2-0-kodirovshhika-dlya-formata-video-av1","title":{"rendered":"Shp\u00ebrthimi i SVT-AV1 4.2.0, koduesit p\u00ebr formatin e videos AV1","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"
\u00cbsht\u00eb publikuar dalja e bibliotek\u00ebs SVT-AV1 4.2.0 (Scalable Video Technology AV1) me implementime t\u00eb koduesit dhe dekoduesit t\u00eb formatit t\u00eb kodimit t\u00eb videos AV1, p\u00ebr t\u00eb cilat shpejt\u00ebsia \u00ebsht\u00eb p\u00ebrshpejtuar duke p\u00ebrdorur zgjerimet q\u00eb ndodhen n\u00eb procesor\u00ebt x86_64 dhe ARM p\u00ebr paralelizimin e llogaritjeve n\u00eb harduer. Projekti \u00ebsht\u00eb krijuar nga kompania Intel n\u00eb partneritet me Netflix me q\u00ebllimin p\u00ebr t\u00eb arritur nj\u00eb nivel performanc\u00eb t\u00eb p\u00ebrshtatsh\u00ebm p\u00ebr rikodimin e videos n\u00eb koh\u00eb reale dhe p\u00ebr aplikim n\u00eb sh\u00ebrbimet q\u00eb ofrojn\u00eb video me k\u00ebrkes\u00eb (VOD). Aktualisht, zhvillimi po vazhdon n\u00ebn mbik\u00ebqyrjen e aleanc\u00ebs Open Media (AOMedia), e cila drejton zhvillimin e formatit t\u00eb kodimit t\u00eb videos AV1. M\u00eb par\u00eb, projekti \u00ebsht\u00eb zhvilluar n\u00eb kuadrin e projektit OpenVisualCloud, i cili gjithashtu zhvillon koduesit SVT-HEVC dhe SVT-VP9. Kodi shp\u00ebrndahet n\u00ebn licenc\u00ebn BSD. <\/p>\n
SVT-AV1 mund t\u00eb nd\u00ebrtohet p\u00ebr sisteme t\u00eb bazuara n\u00eb \u00e7do arkitektur\u00eb p\u00ebr t\u00eb cilat ekziston nj\u00eb kompilator me mb\u00ebshtetje p\u00ebr standardin C99, por performanca m\u00eb e mir\u00eb arrihet n\u00eb sistemet x86_64, p\u00ebr t\u00eb cilat aplikohet optimizimi n\u00eb assembler bazuar n\u00eb instrukcionet SIMD (preferohet q\u00eb CPU t\u00eb ket\u00eb mb\u00ebshtetje p\u00ebr AVX2, por si minimum \u00ebsht\u00eb e mjaftueshme edhe SSE2). Konsumi i memories varet nga numri i b\u00ebrthameve t\u00eb procesorit t\u00eb angazhuara gjat\u00eb kodimit, t\u00eb cilat rregullohen me opsionin «—lp». P\u00ebr shkak t\u00eb nd\u00ebrlikimit t\u00eb algoritmeve t\u00eb p\u00ebrdorura n\u00eb AV1, kodimi i k\u00ebtij formati k\u00ebrkon ndjesh\u00ebm m\u00eb shum\u00eb burime se formate t\u00eb tjera. <\/p>\n
Nd\u00ebr ndryshimet n\u00eb versionin e ri t\u00eb SVT-AV1: <\/p>\n
\n
\u00ebsht\u00eb realizuar nj\u00eb mod t\u00eb ri optimizimi TUNE-VMAF, q\u00eb mund\u00ebson kursimin deri n\u00eb 15% t\u00eb bitrate-it me t\u00eb nj\u00ebjtin nivel cil\u00ebsie vizuale sipas metrikes VMAF, e cila vler\u00ebson perceptimin njer\u00ebzor, dhe me nj\u00eb ulje t\u00eb vog\u00ebl n\u00eb cil\u00ebsin\u00eb sipas metrikes PSNR, e cila vler\u00ebson sakt\u00ebsin\u00eb e riprodhimit t\u00eb t\u00eb dh\u00ebnave origjinale n\u00eb nivelin e piksel\u00ebve.\n
\u00ebsht\u00eb realizuar nj\u00eb mod kodimi me nj\u00eb thread, i cili mund t\u00eb p\u00ebrdoret me modin e qasjes s\u00eb rast\u00ebsishme (Random Access).\n
N\u00eb presetet M3-M5, \u00ebsht\u00eb optimizuar kompromisi mes cil\u00ebsis\u00eb\/shpejt\u00ebsis\u00eb gjat\u00eb kodimit n\u00eb modin e qasjes s\u00eb rast\u00ebsishme.\n
Jan\u00eb shtuar opsione t\u00eb reja n\u00eb komand\u00ebn e linj\u00ebs: —cqp, —enable-intrabc, —hbd-mds, —enable-kf-tf.\n
Shtuar mb\u00ebshtetje p\u00ebr output n\u00eb formatin raw OBU p\u00ebrve\u00e7 formatit IVF q\u00eb mb\u00ebshtetej m\u00eb par\u00eb.\n
\u00cbsht\u00eb optimizuar kompromisi cil\u00ebsi\/shpejt\u00ebsi gjat\u00eb kodimit n\u00eb modin «—rtc» (Komunikime n\u00eb Kohe Reale), si dhe \u00ebsht\u00eb rritur efikasiteti i kompresimit dhe \u00ebsht\u00eb ulur konsumi i memories gjat\u00eb kodimit n\u00eb rezolucione t\u00eb ul\u00ebta. Jan\u00eb shtuar parameter\u00ebt «—max-intra-bitrate-pct» dhe «—max-inter-bitrate-pct».\n
\u00cbsht\u00eb shtuar nj\u00eb mod i kodimit n\u00eb koh\u00eb reale me bitrate t\u00eb vazhduesh\u00ebm — CBR (Bitrate t\u00eb Vazhduesh\u00ebm), t\u00eb realizuar me p\u00ebrdorimin e filtrit Kalman dhe q\u00eb mb\u00ebshtet azhurnimin ciklik me p\u00ebrhapjen e ndryshimeve komplekse p\u00ebrmes disa kadrove dhe ri-kodimin e kadrove me cil\u00ebsi m\u00eb t\u00eb ul\u00ebt, n\u00ebse ato nuk bien brenda bitrate t\u00eb caktuar.\n
N\u00eb modin «—rtc» \u00ebsht\u00eb shtuar mb\u00ebshtetje p\u00ebr ndryshimin dinamik t\u00eb cil\u00ebsimeve t\u00eb bitrate dhe presets n\u00eb m\u00ebnyr\u00eb spontane.\n
Shtuar API p\u00ebr menaxhimin e kadrove mb\u00ebshtet\u00ebse, i cili mb\u00ebshtet LTR (Referenc\u00eb afatgjat\u00eb) dhe struktur\u00ebn me dy nivele t\u00eb grupeve t\u00eb kadrove mb\u00ebshtet\u00ebse (RPS, Grupi i Imazheve Referenc\u00eb).\n
Rishqyrtuar implementimi i kodimit t\u00eb entropis\u00eb, ku \u00ebsht\u00eb e mund\u00ebsuar shkrimi direkt n\u00eb buferin e segmenteve (tile-buffer), \u00ebsht\u00eb thjeshtuar koduesi aritmetik dhe \u00ebsht\u00eb realizuar prerja e t\u00eb dh\u00ebnave t\u00eb mbetura pas kualtizimit.\n
Shtuar optimizime t\u00eb reja CDEF (Filtri i P\u00ebrmir\u00ebsimit t\u00eb Drejt\u00eb t\u00eb Kufizuar) p\u00ebr p\u00ebrjashtimin e mjegull\u00ebs n\u00eb zonat e kalimit t\u00eb kontrastit, MD (Vendosja e Modings) dhe ME (Vler\u00ebsimi i L\u00ebvizjes) p\u00ebr p\u00ebrpunim t\u00eb shpejt\u00eb t\u00eb objekteve statike.\n
Optimizuar identifikimi i pranis\u00eb s\u00eb imazhit n\u00eb video me p\u00ebrmbajtje ekrani p\u00ebr t\u00eb aplikuar metoda specifike kodimi p\u00ebr ruajtjen e qart\u00ebsis\u00eb s\u00eb tekstit.\n
\u00cbsht\u00eb shtuar nj\u00eb opsion nd\u00ebrtimi ENABLE_STACK_PROTECTOR p\u00ebr t\u00eb \u00e7aktivizuar flagun «-fstack-protector», duke lejuar eliminimin e kostove p\u00ebr kontrollin e mbingarkes\u00ebs n\u00eb stack.\n
Shtuar optimizime shtes\u00eb p\u00ebr arkitektur\u00ebn ARM dhe aktivizuar instruksionet e zgjeruara NEON p\u00ebr t\u00eb p\u00ebrshpejtuar llogaritjet. <\/ul>\n