by rigaya

日本語版はこちら＞＞

This software is meant to investigate performance and image quality of HW encoder (QSV) of Intel. There are 2 types of software developed, one is command line version that runs independently, and the nother is a output plug-in of Aviutl.

• QSVEncC.exe ... Command line version supporting transcoding.
• QSVEnc.auo ... Output plugin for Aviutl.

rigayaの日記兼メモ帳＞＞
github releases

Install instructions for Windows and Linux

Build instructions for Windows and Linux

Windows 10/11 (x86/x64)
AviUtl 1.00/1.10 (QSVEnc.auo)
AviUtl 2.00 beta19 or later (QSVEnc.auo2)

Debian/Ubuntu (QSVEncC)
Fedora (QSVEncC)
Requires Broadwell CPU or later.
It may be possible to run on other distributions (not tested).

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND.

Option list and details of QSVEncC

• Encoding using QuickSyncVideo

  • H.264/AVC
  • H.265/HEVC (8bit/10bit)
  • MPEG2
  • VP9 (8bit/10bit)
  • AV1 (8bit/10bit)

• Encode mode of QuickSyncVideo

  • CQP (Fixed Quantization)
  • CBR (Constant bitrate)
  • VBR (Variable bitrate)
  • AVBR (Adaptive Variable bitrate)
  • QVBR (Quality based Variable bitrate)
  • LA (Lookahead mode)
  • LA-HRD (HRD compatible Lookahead mode)
  • ICQ (Constant Quality)
  • LA-ICQ (Constant Quality with Lookahead)
  • VCM (Video Conference mode)

• Quality preset of 7 steps

• Interlaced encoding (by PAFF)

• Supports setting of codec profile & level, SAR, colormatrix, maxbitrate, GOP len, etc...

• Parallel encoding supporting multi GPU

• Calculation of ssim/psnr of the encode

• VPP (Video Pre-Processing) filters

  Category	Filters
  Deinterlace	deinterlace (Media Function), afs, bwdif, yadif, nnedi, rtgmc, kfm, decomb, onnx-deint
  Inverse Telecine / Decimation	rff, ivtc, decimate, mpdecimate
  Noise Reduction	knn, pmd, nlmeans, hqdn3d, smooth, denoise-dct, fft3d, msmooth, degrain, convolution3d, denoise (Media Function), mctf
  Resize	resize (various algorithms, libplacebo), descale
  Edge / Detail Enhancement	unsharp, edgelevel, warpsharp, maa, cas, msharpen, detailsharpen, detail-enhance (Media Function)
  Dehalo / Deringing	dehalo, finedehalo, hqdering, vinverse
  Color Adjustment	tweak, curves, softlight, chromashift, colorfix
  Color Space / HDR	colorspace, libplacebo-tonemapping
  Debanding	deband, libplacebo-deband
  Frame Interpolation	rife-ov, ai-frameinterp
  Other	delogo, subburn, pad, overlay, rotate, transform, stab, deflicker, deblock, image-stab (Media Function), libplacebo-shader, onnx, anime4k-shader

• Supports QSV(hw) decoding
  • MPEG2
  • H.264/AVC
  • HEVC
  • VP8
  • VP9
  • AV1
• Supports various formats such as avs, vpy, y4m, and raw
• Supports demux/muxing using libavformat
• Supports decode using libavcodec

• Audio encoding
• Mux audio and chapter
• afs (Automatic field shift) support

This depends on the version of QSVEnc, the generation of the GPU, and also the GPU driver.

QSVEncC will automatically select a GPU depending on the options used, when there are multiple GPUs available which support QSV. --device option can be used to specify on which GPU to run manually.

1. Select GPU which supports...
   Items below will be checked whether the GPU supports it or not

• Codec, Profile, Level
• Additionally, below items will be checked if specified
  • 10bit depth encoding
  • HW decode

2. If there are multiple GPUs which suports all the items checked in 1. and 2., GPU below will be prefered.

• GPU which has low Video Engine(VE) utilization
• GPU which has low GPU core utilization

The purpose of selecting GPU with lower VE/GPU ulitization is to assign tasks to mulitple GPUs and improve the throughput of the tasks.

Please note that VE and GPU ulitization are check at the initialization pahse of the app, and there are delays in values taken. Therefore, it is likely that the multiple tasks started at the same time to run on the same GPU, and divided into multiple GPUs, even if the options are supported in every GPUs.

• MIT license.
• This software depends on oneVPL, ffmpeg, libplacebo, libhdr10plus, libdovi, libass, tinyxml2, dtl, clRNG, ttmath & Caption2Ass. For these licenses, please see the header part of the corresponding source and license.txt.

Windows ... VC build
Linux ... makefile + gcc/clang build

Character code: UTF-8-BOM
Line feed: CRLF
Indent: blank x4