DHR-動畫研究字幕組

標題: X265測試影片發布帖 (2016/7/30) [打印本頁]
作者: 風影氣旋    時間: 2014-2-21 13:58
標題: X265測試影片發布帖 (2016/7/30)
本帖最後由 風影氣旋 於 2016-7-30 13:50 編輯

鑑於x265還在測試階段  一直放在bt種子裡也不是好辦法
於是打算放網盤丟過來這邊, 還能邊做紀錄
首先是介紹編碼
x265是基於HEVC標準開發的開源編碼器
高效率視頻編碼（High Efficiency Video Coding，簡稱HEVC）是一種視訊壓縮標準，H.264/MPEG-4 AVC的繼任者。
更多詳細可參考 wiki HEVC

紀錄
2013年
11/3   LAV Filters 0.59.0 支持解碼H.265/HEVC
11/6   測試版 超次元戰記 戰機少女 NCOP 1080P
11/11  測試版 東京暗鴉 NCOP 720P

2014年
2/12   FFMPEG支持HEVC編碼
2/21   測試版 少女騎士物語 NCOPED 1080P
3/3     MKV支持H.265/HEVC封裝
3/14   新增 true tears×花咲くいろは×TARI TARI 特別アニメーション「ユートピア」HEVC
5/15   新增噬血狂襲 NCOPED 1080P / encode x265－1.0.5
※因掃黃打非騰訊微雲關閉影片上傳 另傳網盤分流
8/11   新增噬血狂襲 NCOPED 1080P test2、果然我的青春戀愛喜劇搞錯了 NCOPED 1080P、LoveLive！2nd Season NCOP 1080P、請問您今天要來點兔子嗎？NCOPED 1080P / encode x265－1.2.496
9/15 AMD顯卡專用 OpenCL版HEVC/H.265解碼器 http://xhevc.com/cn/downloads/downloadCenter.jsp 、 新增噬血狂襲 NCOPED2 1080P test

2015年
1/06 新增 甘城輝煌樂園救世主 NCOPED 1080P TEST(YUV422P8、YUV420P10) X265版本1.4+244
1/16 Intel發佈新內顯驅動版本號15.36.14.4080 支持Haswell/Broadwell CPU、HEVC硬解（據說是混合...？）、4K解碼
https://communities.intel.com/thread/59216
1/23 NVIDIA發售全世界首款支持HEVC純硬解（8bit+10bit) 4K解碼的GeForce GTX 960
2/13 LAV Filters 0.64.0 (released 2015/02/13)
2/21 dropbox http://1drv.ms/1vnE4BL 數個測試更新
10/9 x265 1.8版 支持12bit
11/11 偽戀BD NCOPED dropbox
11/12 ROCK IN JAPAN FESTIVAL.2015_LiSA LIVE

2016年
1/29 x265 1.9版
4/23 無彩限的幻影世界BD NCOPED
7/13 x265 2.0版

測試影片下載
超次元戰記 戰機少女 NCOP 1080P
東京暗鴉 NCOP 720P
少女騎士物語 NCOP 1080P
少女騎士物語 NCED 1080P
true tears×花咲くいろは×TARI TARI 特別アニメーション「ユートピア」HEVC
噬血狂襲 NCOP 1080P
噬血狂襲 NCED 1080P

Mega x265_test   
※改以資料夾分享

百度備用載點
链接：http://pan.baidu.com/s/1i3BjD9z 密码：57lm

dropboxNEW

不定時更新

動漫花園連結

作者: mrwu1990    時間: 2014-8-13 20:55
作者: stevenweng09    時間: 2015-1-29 02:11
手機/平板 2015的 高通810 支援4K HEVC 硬解
還有4K HEVC 錄影 編碼

HEVC可以開使普及了www

BTW :4K Blu_ray Disc 也要採用HEVC編碼了
作者: stevenweng09    時間: 2015-1-29 02:16
本帖最後由 stevenweng09 於 2015-1-29 02:53 編輯

請問 果青 還有 strike the blood
是色彩用8bit還是 10bit (Hi10P)???

作者: 風影氣旋    時間: 2015-2-21 22:37
stevenweng09 發表於 2015-1-29 02:16
請問 果青 還有 strike the blood
是色彩用8bit還是 10bit (Hi10P)???

去年的測試幾乎是8bit 那時10bit還有bug...

作者: 風影氣旋    時間: 2015-2-21 22:39
本帖最後由 風影氣旋 於 2015-2-21 22:42 編輯

補充：引用LAV Filters作者的話

LAV Filters 0.64.0 (released 2015/02/13)
General
- NEW: Individual LAV Filters can now be blacklisted through the registry by creating a key with the executable name in HKCU\Software\LAV\<Filter>\Blacklist

LAV Splitter
- NEW: Support for RTMP using rtmpdump-style parameter syntax
- NEW: Support for Opus-in-TS
- Fixed: The language reported for audio/subtitle streams on some Blu-rays could be wrong

LAV Video
- NEW: HEVC Main10 decoding in DXVA2 Copy-Back mode
- Faster: DXVA2 Copy-Back in direct output mode uses up to 50% less CPU and performance is improved accordingly
- Fixed: H.264 streams with only DTS timestamps would play out of sync with DXVA2
- Fixed: DXVA2 could crash on some MPEG-2 streams
- Fixed: Improved compatibility with a few HEVC streams when using DXVA2

DXVA2 HEVC Main10 (10-bit) support

LAV Video now supports 10-bit HEVC decoding through DXVA2, on hardware which supports the Main10 profile!
It has been developed and tested on the NVIDIA GTX 960, which is the first desktop GPU with a full hardware HEVC decoder, and it works quite nice and fast (4k @ 100-120 fps)

At this time I have no information if it also works on Intels Main10 decoding on the Broadwell iGPU, since I do not have access to such hardware.

Note that at this time 10-bit decoding is limited to DXVA2 Copy-Back mode, because the video renderers did not want to cooperate in Native 10-bit mode.

DXVA2 Copy-Back "Direct" mode

To reduce the CPU overhead and the speed limitations of the DXVA2 Copy-Back mode, LAV Video 0.64 introduces a new "Direct" mode.
Direct Mode reduces the CPU overhead by up to 50%, resulting in significantly reduced CPU usage figures.

The concept is simple: When possible, LAV Video will do one processing step less, and try to copy the video image from the GPU to the video renderer as directly as possible.
This has a few limitations however. It only works in specifically designed output paths, specifically NV12 decoding -> NV12 output, P010 decoding -> P010 output, and P010 decoding -> NV12 output.
In addition, direct mode cannot be used when YADIF deinterlacing is required.

Direct Mode is not an option, it'll simply turn on when its possible to be used, and turn off when not (ie. with YADIF or when an unsupported conversion is required).

More...

Everything else should be pretty self-explanatory. Nothing too fancy. RTMP is now officially supported using the same parameter syntax as librtmp/rtmpdump use. The feature was contributed by the MPC-HC authors.
Opus in MPEG-TS streams can now be demuxed and decoded. To make this possible, Opus is now decoded by the "native" ffmpeg Opus decoder, and no longer libopus.

Soon..

I have some plans to work on a few of the long standing feature requests and a few of my own ideas very soon, so a new feature push may not be too far out.

In any case, have fun!

PS:
Impatient people, don't even allow an hour to post the versions everywhere and write a post, jeez.



作者: 風影氣旋    時間: 2015-11-11 14:34
x265 version 1.8 has been released. This release supports 12bit input depths, a large amount of AVX2 optimizations, entropy coding optimizations, as well as new quality features.
Full documentation is available at http://x265.readthedocs.org/en/stable/
================== API Changes =====================
• Experimental support for Main12 is now enabled. Partial assembly support exists.
• Main12 and Intra/Still picture profiles are now supported. Still picture profile is detected based on x265_param::totalFrames.
• Three classes of encoding statistics are now available through the API.
o x265_stats - contains encoding statistics, available through x265_encoder_get_stats()
o x265_frame_stats and x265_cu_stats - contains frame encoding statistics, available through recon x265_picture
• --csv
o x265_encoder_log() is now deprecated
o x265_param::csvfn is also deprecated
• --log-level now controls only console logging, frame level console logging has been removed.
• Support added for new color transfer characteristic ARIB STD-B67
=================== New Features ===================
• limit-refs
o This feature limits the references analysed for individual CUS.
o Provides a nice tradeoff between efficiency and performance.
• aq-mode 3
o A new aq-mode that provides additional biasing for low-light conditions.
• An improved scene cut detection logic that allows ratecontrol to manage visual quality at fade-ins and fade-outs better.
================ Preset and Tune Options ===============
• tune grain
o Increases psyRdoq strength to 10.0, and rdoq-level to 2.
• qg-size
o Default value changed to 32.
作者: 風影氣旋    時間: 2016-2-11 00:19
We're pleased to announce the release of x265 version 1.9.

This release supports many new features as well as additional assembly optimizations for Main12, intra prediction and SAO. Recently added features lookahead-slices, limit-refs and limit-modes have been enabled by default in the supported presets.

Full documentation is available at http://x265.readthedocs.org/en/latest/

================= New Features ====================

- Quant offsets: This feature allows block level quantization offsets to be specified for every frame. An API-only feature.
- --intra-refresh: Keyframes can be replaced by a moving column of intra blocks in non-keyframes.
- --limit-modes: Intelligently restricts mode analysis.
- --max-luma and --min-luma for luma clipping, optional for HDR use-cases
- Emergency denoising is now enabled by default in very low bitrate, VBV encodes
================== API Changes ==================

- x265_frame_stats returns many additional fields: maxCLL, maxFALL, residual energy, scenecut and latency logging
- --qpfile now supports frametype 'K"
- x265 now allows CRF ratecontrol in pass N (N greater than or equal to 2)
- Chroma subsampling format YUV 4:0:0 is now fully supported and tested

============= Presets and Performance ===============

- Recently added features lookahead-slices, limit-modes, limit-refs have been enabled by default for applicable presets.
- The default psy-rd strength has been increased to 2.0
- Multi-socket machines now use a single pool of threads that can work cross-socket.
作者: 風影氣旋    時間: 2016-7-30 13:51
本帖最後由 風影氣旋 於 2016-7-30 13:53 編輯

x265 version 2.0 has been released. This release supports many new features as well as support for ARM assembly optimizations for most basic pixel and ME operations, as well as SAO cleanups and a fully tested reconfigure functionality.

Full documentation is available at http://x265.readthedocs.org/en/stable/

==================== New Features====================

• uhd-bd: Enforce Ultra-HD Blu-ray Disc parameters (overrides any other settings)
• rskip: Enables skipping recursion to analyze lower CU sizes using heuristics at different rd-levels. Provides good visual quality gains at the highest quality presets.
• rc-grain: Enables a new rate control mode specifically for grainy content. Strictly prevents QP oscillations within and between frames to avoid grain fluctuations.
• tune grain: A fully refactored and improved option to encode film grain content including QP control as well as analysis options.
• asm: ARM assembly is now enabled by default, native or cross compiled builds supported on armv6 and later systems.
============ API and Key Behavior Changes ==============
• x265_rc_stats added to x265_picture, containing all RC decision points for that frame
• PTL: high tier is now allowed by default, chosen only if necessary
• multi-pass: First pass now uses slow-firstpass by default, enabling better RC decisions in future passes
• pools: fix behaviour on multi-socketed Windows systems, provide more flexibility in determining thread and pool counts
• ABR: improve bits allocation in the first few frames, abr reset, vbv and cutree improved
====================== Misc =========================
• An SSIM calculation bug was corrected



歡迎光臨 DHR-動畫研究字幕組 (https://dhr.moe/) Powered by Discuz! X3.1