6.1.1

Note

November 9, 2024: A build has been added for Ubuntu 22.04. The Ubuntu 24.04 build has also been updated to add some missing dependencies.

Added

• Time remaining, and processing speed to the status bar.

Fixed

• Stream mapping for cases where the video stream is not the first stream in the input file (#1217).
• The encoder using the wrong color space for the output video.

6.1.0

Note

The Linux binary release has been replaced with a Debian .deb package built for Ubuntu 24.04.

Added

• A better timer that gets paused when the processing is paused.
• Detection for the validity of the provided GPU ID.
• The --listgpus option to list available Vulkan GPU devices.
• Vulkan device selection for libplacebo.
• Status bar and processing statistics. (Video2X Qt6)

Fixed

• Wide character string paths support on Windows systems without UTF-8 suppoprt enabled.

Changed

• Automatically detect if options colorspace and range are supported by the buffer filter.
• Resource file missing error messages.
• Rewritten the CLI with C++.

6.0.0

This is the official release of Video2X version 6.0.0.

TL;DR: Version 6.0.0 is a complete rewrite of the Video2X project in C/C++, featuring a faster, more efficient architecture, cross-platform support, vastly improved output quality, and a new GUI and installer for easy setup on Windows.

Note

October 29, 2024: The Windows installer has been updated to automatically install the VC++ Runtime if required and to check for Vulkan support, warning the user if the Vulkan runtime is missing.

Complete Summary of Changes in 6.0.0

Version 6.0.0 is a complete rewrite of this project in C/C++. It:

• genuinely works this time, with much less hassle compared to the 5.0.0 beta;
• is blazing fast, thanks to the new optimized pipeline and the efficiency of C/C++;
• is cross-platform, available now for both Windows and Linux;
• offers significantly better output quality with Anime4K v4 and RealESRGAN;
• supports Anime4K v4 and all custom MPV-compatible GLSL shaders;
• includes support for RealESRGAN (all three models) via ncnn and Vulkan;
• requires zero additional disk space during processing, just space for the final output; and
• exports a standard C function for easy integration into other projects! (documentations are on the way)

The graphical user interface currently supports the following languages:

• English (United States)
• 简体中文（中国）
• 日本語（日本）
• Português (Portugal)

Why the 5.0.0 branch was abandoned:

• Wrapped C++ libraries for Python were too complex to build for cross-platform distribution.
• Certain wrapped C++ libraries exhibited unexpected behaviors.
• Running FFmpeg through commands and piping data via stdin/stdout proved inefficient.
• Native C/C++ binaries are significantly smaller and more efficient, leveraging the inherent speed of machine code, compiler optimizations, and link-time optimization.

All Notable Changes From 6.0.0-beta.5

• feat: added all Anime4K v4 modes (A, B, C, A+A, B+B, C+A, GAN)
• fix: fixed the issue where custom GLSL shader paths are formatted incorrectly on Windows
• fix: fixed the issue where the output video has a different advertised frame rate than the original video
• fix: fixed missing and incorrect command line arguments

6.0.0-beta.5

Without major issues and bugs, this should be the last beta build before the official release of 6.0.0.

All Notable Changes From 6.0.0-beta.4

• feat: added more modes for Anime4K v4 (A, B, C, A+A, B+B, C+A)
• feat: added the about dialog and a report bugs link
• fix: fixed the issue where the user can select an invalid scaling ratio for the selected RealESRGAN model
• fix: fixed the bug where non-English systems cannot switch to the English locale

libplacebo now comes with all Anime4K v4 modes:

A short description for each mode is available in the tooltip:

The about page has been implemented:

6.0.0-beta.4

All Notable Changes From 6.0.0-beta.3

• fix: fixed a memory leak in the encoder
• feat: use OpenCV to read the total number of frames for videos that don't have nb_frames in their metadata
• feat: added Japanese (Japan) translation for the GUI (thanks @reindex-ot)
• feat: added Portuguese (Portugal) translation for the GUI (thanks @ruiramalhoofc)
• feat: improved the looks of the progress bar

A new progress bar:

Japanese (Japan) translation:

Portuguese (Portugal) translation:

6.0.0-beta.3

All Notable Changes From 6.0.0-beta.2

• feat: switched logging to spdlog with log levels
• docs: fixed docs for libvideo2x functions
• feat: organized header files
• feat: GUI translations for Simplified Chinese

You can now open a log console from the GUI. libvideo2x's internal logging has also been switched to proper logging with spdlog from printf.

Translation for Simplified Chinese has been added to the UI. This includes all UI elements, pop-up messages, and tooltips.

6.0.0-beta.2

Note

October 10, 2024: Added an installer for Video2X Qt6 for Windows.

All Notable Changes From 6.0.0-beta.1

• feat: added support for copying audio and subtitle streams
• feat: added hwaccel for encoder and decoder (unstable)
• feat: added benchmark mode (--benchmark)
• feat: added pause and abort for CLI (SPACE to pause, q to abort)
• chore: updated default preset and crf settings

Video2X 6.0.0 now has a new Qt6 GUI:

Pause and abort for CLI:

Copying audio & subtitle streams:

6.0.0-beta.1

A complete rewrite of the project in C/C++.

The container image is available here.

Video2X 5.0.0-beta5

• Added support for RealCUGAN ncnn Vulkan
• Added pause function
  • Use the global hotkey Ctrl+Alt+V or send SIGUSR1 to the main process to pause
• Fixed various problems with the progress bar

The container has also been patched so it will run fine on both AMD and NVIDIA GPUs. The required Docker arguments have changed for AMD GPUs. See the Wiki page for more information.

Video2X 5.0.0-beta4

• Enabled audio, subtitle, data, and attachment streams copying support
• Fixed a bug in argument parsing