Wowza Community

Wowza Streaming Engine Release v4.8.22 is here!


This is a big one, we are pleased to present Wowza Streaming Engine v4.8.22!! The main highlights of this release are:

Support for AMD Xilinx Alveo U30 in preview mode
With Wowza Streaming Engine 4.8.22, you can transcode Wowza Streaming Engine with an AMD Xilinx Alveo U30 accelerator card in a preview mode in addition to using the CPU, NVIDIA cards, or Intel QuickSync Video.
This enhancement allows users to seamlessly leverage the hardware-accelerated features of Xilinx video codec units, enabling high-density real-time transcoding for live-streaming video.

MainConcept replacement of Beamr for H.264 and HEVC/H.265 transcoding
This release sunsets support for Beamr H.264 and HEVC/H.265 transcoding and updates the default encoding and decoding implementations for H.264 and HEVC/H.265 video to MainConcept.
The move to MainConcept delivers improved quality, speed, and performance. If you update to Wowza Streaming Engine 4.8.22 from an earlier version, existing Transcoder templates that use Default as the encoding or decoding implementation will use MainConcept after the upgrade. You will need to update your encoding configuration for MainConcept if it uses any advanced Beamr parameters.

Upgrade to CUDA 12
With Wowza Streaming Engine release 4.8.22, we’ve upgraded support to CUDA 12 for NVIDIA GPU-accelerated live stream transcoding.
This upgrade supports using the latest NVIDIA drivers for transcoding at greater scale and speed. The NVIDIA microarchitecture of your hardware must support CUDA 12. NVIDIA driver versions must be at least 525.60.13 for Linux and 527.41 for Windows. Any previous or unsupported drivers will cause Wowza Streaming Engine to revert back to default CPU transcoding.

There’s A LOT here, for the full list of updates please see the documentation below:

Release notes

Known issues

Article Updates
Introduction to AMD Xilinx acceleration
Server specs for AMD Xilinx acceleration
Limitations of AMD Xilinx acceleration
Configure an AMD Xilinx accelerated transcoding
Upgrade to CUDA 12

Beamr to MainConcept Updates

1 Like

Why the switch back from Beamr to Mainconcept Transcoder?

The new Mainconcept Implemenatiion shows better Picture-Quality since the old. But i cant see much better Live-Transcoding Quality compared with Beamr. And Mainconcept needs significant more CPU-Power then Beamr.

Beamr needs on a DELL AMD Epyc 7302 (16/32 Core, 3Ghz, 64GB RAM on Windows Server 2016) dedicated Machine with 1 Stream (720p50 @5Mbit, 3 Renditions 540p,480p,360p and original 720p as Passtrough) near 10-15% CPU-Power.
Mainconcept needs on the same Machine with the same Stream-Setup 20-25% CPU-Power. Same on Intel Machines.