Real-Time UHD Video Processing & Audio DSP

Audio and video systems are in constant pursuit of providing a more immersive and realistic experience with all the accuracy and fidelity as originally captured. Whether this is higher resolution, higher dynamic range and higher frame rate image and video processing, or full surround sound with object-based audio, the trends require much more processing capabilities. XAMD platforms offer a combination of embedded software processing where it makes sense, along with high-performance hardware accelerated DSP and memory to process real-time multichannel UHD video algorithms or multichannel audio effects. AMD also provides many of the fundamental building blocks used in typical professional multimedia pipelines, getting you to prototype and deployment much quicker, whilst leaving plenty of room for differentiation.





Scalability to 4K, 8K and beyond to create a more immersive experience



Higher frame rates to reduce blurring of high-speed movements



Wider color gamuts such as Rec.2020 for more natural color representation



Higher dynamic range for improved darks and lights



Arbitrary warp and rotation for keystone and fisheye correction



Increasing number of audio channels and new sound fields for total audio immersion


Up to 8K scaling, deinterlacing, chroma resampling and color space conversion

The Video Processing Subsystem (VPSS) from AMD is a collection of video processing IP subcores, bundled together in hardware and software, abstracting a typical real-time 8K 60fps video processing pipeline. From scaling to deinterlacing, color space conversion and chroma resampling, this highly configurable and scalable subsystem can kickstart your next video design. Provided free in Vivado™, the VPSS provides users an out of the box and ready to use video processing core which can accelerate development in a wide range of broadcast and AV applications.



Alpha blending up to seventeen layers of video and graphics

The AMD LogiCORE™ IP Video Mixer core provides a fully flexible video processing block for alpha blending and compositing multiple video and/or graphics layers. Support for up to seventeen layers (one main layer and sixteen overlay layers), with an optional logo layer, using a combination of video inputs from either memory or streaming video cores is provided. Provided free in Vivado™, the AMDVideo Mixer offers scalability from simple on-screen display (OSD) for user interfacing, to professional quality compositing, and can be used in multiviewers, switchers and video conferencing systems.



Create up to eight scaled outputs from single or multiple video inputs

The AMD LogiCORE IP Multi-Scaler generates up to eight scaled output images from a single or multiple (up to eight) external video and/or graphics sources. It is a memory interface-based core which is highly configurable and supports in-system programmability through a comprehensive register interface to control it. Supporting spatial resolutions from 64x64 to 8192x4320, the multi-scaler is an ideal solution to simultaneously creating multiple sizes of outputs for video walls, multiviewers and archival systems.



Enabling lens correction, projection mapping and creative video wall presentation

The AMD Video Warp Processor Subsystem supports real-time image distortions such as barrel, pincushion, keystone, arbitrary distortion, rotation, translation and scaling. Attractively priced and supporting up to 4Kp60 video with low latency, the Video Warp Processor is ideal for applications across Broadcast & Pro AV. It can be used in applications such as collaboration to correct wide-angle lenses, or projector products for keystone correction and mapping to curved or irregular surfaces, and in video walls to create eye-catching monitor arrangements.


Perform real-time HDR conversion without compromising artistic intent

Many of the AMD video IP cores and codec reference designs recognise HDR metadata in HDR10 and HLG formats. In many cases, this is simply a case of detecting static or dynamic HDR formats and passing them through to the next stage of the media processing chain. But many workflows need to deal with both SDR and HDR content and need a way to convert between the two – and this is not a straight-forward operation as it can easily change the original artistic intent if not handled properly.

AMD partner b<>com have vast experience in HDR workflows. Their Sublima SDR-HDR converter IP offers all the benefits of real-time frame-by-frame adaptive conversion techniques, without any need for manual adjustment. Based on an intelligent algorithm, this product guarantees an optimal conversion regardless of the video content. Despite the adaptive conversion, it does not require the use of metadata, and yet is still able to guarantee a visually lossless round trip.



High performance, high density and low latency audio DSP

AMD platforms are unrivalled in providing the digital signal processing (DSP) performance required in audio processing, interfacing, compression, embedding and conversion. The inherent parallelism of AMD architectures means that many channels of audio can be processed together using very efficient resources. AMD Vitis development tools support high level languages such as C++, so audio engineers can rapidly integrate audio functionality into their products, offering differentiation, channel density and fast time-to-market.



The latest AV connectivity IP bundled together in a low-cost package

We know the cost of IP can quickly mount up, especially for complex multimedia systems. To make things easier and more cost-effective, AMD has bundled together the common AV connectivity IP and subsystems together in a single purchase, with a site license costing only $10K.

The Video IP Toolbox includes:

This is in addition to the free subsystems included with Vivado™ & Vitis™ tools:

Ask your AMDor distributor sales team for:

  • EF-DI-XP4KVIDEO-SITE (for a site license)
  • EF-DI-XP4KVIDEO-WW (for a worldwide license)