Video Encoding, Codec & Transcoding News

Dolby has acquired a portfolio of VVC standard essential patents from Florida Atlantic University and OP Solutions, while major automakers have abandoned plans for a joint licensing negotiation group following antitrust scrutiny. Additionally, new FRAND guidelines from the Munich Regional Court and a Brazilian court ruling on pool offers are shaping the global landscape for video codec patent licensing.

Researchers have introduced a Generalized Slimmable Framework for Implicit Neural Representations (INR) that allows a single model checkpoint to support multiple bitrates. By utilizing width-configurable layers and Slimmable Conditional Decoder Modulation, the framework reduces multi-rate storage requirements by up to 2.5x while maintaining reconstruction quality across diverse INR backbones.

Amazon's acquisition of MX Player and Dolby's $429 million purchase of GE Licensing's patent portfolio suggest a potential increase in industry momentum for the VVC codec. These developments highlight the ongoing interest in VVC for software-based playback and the strategic importance of codec patent portfolios for major technology players.

MPEG issued a Joint Call for Proposals for next-generation video compression technology exceeding VVC capabilities, targeting a final standard by 2029. The organization also finalized standards for the 9th edition of the ISO base media file format, avatar representation, and enhanced geometry-based point cloud compression.

Interra Systems will showcase updates to its ORION, BATON, and VEGA platforms at SET Expo 2026, including new AI-driven captioning, VVC support, and NDI monitoring. The release aims to address operational challenges for LATAM media organizations managing increased FAST, live sports, and multi-platform streaming workflows.

Microsoft has released .NET 11 Preview 7, which includes significant updates to the runtime and developer SDK. Key features relevant to streaming engineers include tiered compilation for async methods, expanded WebAssembly support, and new HttpContent wrappers for GZip, Brotli, and Zstandard compression.

Vivaro utilized the Appear X20 platform to manage live video delivery for all 104 matches of the 2026 FIFA World Cup across North and South America. The deployment supported both HEVC and JPEG XS codecs to handle varying bandwidth and latency requirements for linear and digital platforms.

Z3 Technology has introduced the ZCube-21, a compact H.265/H.264 encoder supporting up to 4Kp30 resolution with triple-stream output capabilities. Designed for specialized sectors like military and surveillance, the device features locking HDMI, PoE support, and various control protocols including SRT and RTMP.

Qualcomm has released full specifications for its 6nm Snapdragon C chipset, a platform designed for entry-tier PCs that includes a dedicated video processing unit for H.264 and H.265 hardware acceleration. The chipset features an 8-core Kryo CPU and Hexagon NPU, with hardware support for 4K/60 FPS display output, targeting upcoming devices from manufacturers including Acer, ASUS, Lenovo, and HP.

Broadcom has submitted a Linux kernel patch series introducing a standard Performance Monitoring Unit driver for Raspberry Pi SoCs, covering models from the BCM2835 to the BCM2712. The update enables engineers to utilize standard Linux perf tools for monitoring system throughput and hardware-level video processing performance on the VPU and HEVC decoders.

Dolby Laboratories and Beko Elektronik have reached a settlement in patent litigation concerning the Opus audio codec, leading to the cancellation of upcoming Unified Patent Court hearings. The case is notable for establishing that holders of passive standard-essential patents must adhere to the Huawei v. ZTE FRAND negotiation framework and may be considered dominant market players.

NVIDIA released JetPack 7.2.1, which adds support for PyNvVideoCodec 2.2 and new agentic video skills for the Jetson hardware platform. The release allows developers to automate video codec configuration and performance benchmarking for AI-driven robotics and media processing applications.

AMD has identified a synthesis stall issue affecting HLS-based video IP, including the Video Mixer and Multi-Scaler, when running on Windows hosts in Vivado versions 2020.1 through 2021.1. The company provides specific Answer Records to guide users toward software updates or migration to Linux to resolve the compilation hang.

Nokia outlines the ongoing integration of AI into video coding standards including H.266/VVC and H.274/VSEI to enhance compression efficiency and support machine vision. The article details how hybrid pipelines, neural post-filters, and emerging multimodal representations are augmenting traditional block-based coding methods.

M&K Holdings has initiated patent infringement litigation against Snap in the Unified Patent Court, escalating a coordinated enforcement effort by VDP Pool licensors. This action follows similar legal filings by Dolby and ETRI in the US and Europe concerning the licensing of HEVC and AV1 codec technologies.

Dolby Laboratories has acquired a portfolio of 683 intellectual property assets from Florida Atlantic University and OP Solutions. The acquisition includes 16 patents essential to the Versatile Video Coding (VVC/H.266) standard, which significantly improves compression efficiency for streaming and broadcast applications.

Phoronix published benchmarking data comparing video encoding performance between Ubuntu 26.04 LTS and Windows 11 Pro on Intel Xeon 678x hardware. The results indicate that Ubuntu generally leads in SVT-AV1 and OpenAPV encoding tasks, while Windows 11 shows superior performance in certain Kvazaar H.265 scenarios.

Fastvideo has announced a new CUDA-based Motion JPEG codec capable of achieving 5500 fps encoding for Full HD video on NVIDIA RTX 4090 GPUs. The codec is designed for high-speed imaging pipelines that require individual frame accessibility, offering a high-throughput alternative to inter-frame codecs like H.264.

Researchers have proposed a new memory-centered motion estimation framework for VVC/H.266 that utilizes an Adaptive Winograd Coding Unit Partitioning accelerator to reduce computational redundancy. The FPGA-based design demonstrates significant bitrate reduction and quality improvements for high-resolution video compression tasks.

V-Nova SVP Fabio Murra discusses how licensing costs and commercial constraints are increasingly influencing UHD and HDR deployment strategies for streaming operators. He proposes MPEG-5 LCEVC as a solution to optimize costs and quality without requiring a wholesale codec migration.

Microsoft has announced the development of a machine learning video codec designed for NPU-based processing. Concurrently, Akamai has introduced new edge functions to manage entitlement drift in multi-CDN delivery environments.

Beamr Imaging is integrating NVIDIA Video Super Resolution with its content-adaptive bitrate encoding to allow broadcasters to upscale HD live sports to 4K. The combined workflow is designed to reduce delivery stream sizes by up to 50% while maintaining production-controlled quality, with plans for a demonstration at IBC 2026.

A Brazilian court-appointed expert has determined that Disney's HEVC-encoded video streams infringe upon an IBEX patent. The report also concludes that the patent is non-essential to the HEVC standard because it applies to encoding rather than decoding, potentially allowing IBEX to bypass FRAND licensing obligations.

Nvidia RTX 4090 users on Linux are reporting that utilizing the NVENC engine for AV1, H.264, or HEVC encoding triggers immediate and silent system-wide resets. The issue occurs across multiple driver versions and persists without producing kernel error logs, complicating professional streaming and encoding workflows.

Hornets Tech has announced a new series of encoders and decoders for SDI and ST 2110 signals, alongside an IPTV platform, scheduled for demonstration at IBC2026. The hardware is designed to support hybrid broadcast infrastructures by facilitating signal conversion and low-latency wireless RF transmission.

AWS has updated its SDK to include concurrent function execution for Elemental MediaTailor, enabling parallel child function runs to improve latency in streaming pipelines. The update also adds support for various new features across other AWS services, including S3 access points for AWS Backup and updated controls for Amazon Bedrock AgentCore.

Elecard has released StreamEye Studio 2026, an update to its video analysis toolkit that introduces comprehensive AV1 buffer analysis and improved HEVC multithreading capabilities. The release also includes a redesigned Video Quality Estimator and enhanced support for HLS/MPEG-DASH manifest analysis and HDR Vivid metadata.

A developer identified an issue where the Emby media server fails to include necessary VIDEO-RANGE and CODECS metadata in HLS multivariant playlists, preventing native HDR playback on Apple TV. The discussion highlights a technical requirement for correct HLS manifest signaling to trigger Apple's hardware-accelerated decoding and dynamic range switching.

Disney has restored 4K UHD streaming for Disney+ in Germany by implementing a VP9-based workaround following legal injunctions won by InterDigital regarding video codec patents. While 4K resolution is returning to devices like Fire TV Cubes, HDR remains disabled as Disney continues to navigate ongoing global patent litigation.

Modern Streaming Solutions has partnered with VisualOn to offer AI-powered per-title and per-scene video optimization to OTT platforms in the Indian market. The integration allows operators to deploy bitrate-reducing technology via a standalone interface or API to lower storage and CDN egress costs.

The ITU, ISO, and IEC have issued a formal call for proposals for a new video compression standard to succeed VVC. The project aims to develop a new codec by 2029 that incorporates advancements in AI-enabled processing, XR, and machine vision.

A critical vulnerability (CVE-2026-66066) in the Ruby on Rails Active Storage component allows unauthenticated attackers to achieve remote code execution via malicious image uploads. Streaming infrastructure teams utilizing Rails for user-generated content or metadata management must update to patched versions and rotate application credentials to prevent unauthorized system access.

GStreamer 1.28.6 has been released as the final maintenance update for the 1.28 stable series, addressing security vulnerabilities and critical regressions. The update includes fixes for playbin3 and WebRTC, preparing the framework for the upcoming 1.30 release targeted for Q4 2026.

FFmpeg 9.0 has been released with full support for Vulkan-based hardware acceleration, enabling vendor-neutral GPU processing for encoding, decoding, and filtering. The update supports multiple codecs including AV1, HEVC, and H.264 across AMD, NVIDIA, and Intel hardware, reducing dependency on proprietary APIs.

AirFi has launched AirFunnl, a new media delivery stack for inflight entertainment that uses CMAF packaging and AI-enhanced DRM to reduce video file sizes by up to 45%. The solution aims to optimize distribution costs and playback performance for airlines using wireless IFE architectures.

A new research survey examines Neural Video Representations (NVRs) as an alternative to traditional codecs like VVC and HEVC for video compression. The authors evaluate how overfitting small neural networks to specific video sequences can potentially lower decoding complexity and energy consumption for mobile streaming applications.

This technical article examines common caching patterns including cache-aside, write-through, and write-behind, providing implementation examples for Spring Boot 3.x and Redis. It also addresses critical failure modes such as thundering herds and cache penetration, offering strategies like single-flight locking and negative caching to improve system resilience.

Samsung and Prime Video have announced the global launch of HDR10+ ADVANCED, a new HDR standard featuring scene-by-scene metadata for enhanced brightness and AI-driven motion processing. The feature debuts in August 2026 on Samsung's 2026 television lineup for select Prime Video titles.

Appear is launching the ECx400 compression module for its X10 and X20 platforms, alongside new VX Hybrid Cloud Gateway capabilities, at IBC2026. The new hardware module supports JPEG XS, HEVC, and AVC, integrating with Appear's software-defined VX platform to facilitate hybrid ground-to-cloud live production workflows.

The FFmpeg project has released version 9.0, codenamed 'Lei,' introducing significant technical updates including LCEVC track muxing, APV Vulkan hardware acceleration, and an ONNX Runtime DNN backend. The update also adds support for ProRes RAW VideoToolbox acceleration and includes a new bitstream filter for Dolby Vision multi-layer HEVC.

An Emby developer confirmed that the platform's media server has a hardcoded audio transcoding limit of 44.1 kHz. This limitation persists even when users request higher sample rates, potentially impacting workflows requiring hardware compatibility with 48 kHz sources.

This article provides a historical and technical overview of the MP3 audio compression standard, detailing how psychoacoustic modeling and auditory masking reduce file sizes while maintaining perceived audio quality. It explains the standard's origins at the Fraunhofer Institute and its role within the MPEG-1 standard.

Chromium has implemented a change to the Windows H.264 screen-sharing encoder that caps the quantization parameter at 35 to improve text legibility. The update prioritizes maintaining image quality over frame rate during periods of limited network bandwidth, impacting browser-based communication tools.

Microsoft has open-sourced its Machine Learning Video Codec (MLVC) under the MIT License, which is designed for real-time video conferencing on consumer-grade NPUs. The codec utilizes a scale-sharing hyperprior mechanism to resolve cross-platform floating-point drift and is currently undergoing production testing within Microsoft Teams.

Users report significant visual artifacts when utilizing NVIDIA's H.264 Vulkan Video hardware decoder on Linux systems with recent GPU drivers. The issue appears specific to the H.264 implementation under the Vulkan API, as H.265, AV1, and CUDA-based decoding remain unaffected.

AWS has updated its Go SDK to version 2026-07-31.2, which includes new graphic composition capabilities for AWS Elemental Inference on cropped video outputs. The release also introduces transcript formatting improvements for Amazon Transcribe and mid-conversation tool change support for Amazon Bedrock.

NVIDIA has released Video Codec SDK 13.1, which introduces AV1 hierarchical reference mode supporting up to 31 B-frames and a new zero-copy transcode pipeline for improved performance. The update also features improved decoding statistics for analytics, frame-accurate seeking for editing, and an official Docker-based development environment.