Manifold technologies, the Germany-based provider of high-performance cloud infrastructure for live broadcast production, will demonstrate support for 400GbE COTS FPGA accelerator cards at the 2026 NAB Show in Las Vegas (Booth C.1808). The development marks a significant leap in processing density, enabling broadcasters and production companies to handle massive scale with minimal hardware footprint.
Manifold’s service-oriented manifold CLOUD platform delivers broadcast-grade processing, with multiviewers, UDX conversion and graphics insertion – on commercial off-the-shelf (COTS) FPGA accelerator cards. By decoupling processing from proprietary hardware, manifold CLOUD gives users the freedom to choose the acceleration that suits their workflow, supporting cards from BittWare, ProDesign, and other leading vendors. The platform is also optimised to run on technology partner arkona’s AT300 FPGA acceleration cards and works seamlessly with arkona’s BLADE//runner infrastructure – which itself provides deterministic signal gatewaying, frame synchronisation, vision mixing and advanced audio processing. Together, they form a complete, software-defined production ecosystem of remarkable power and scalability.
The addition of 400GbE COTS FPGA support to the manifold CLOUD platform transforms what’s possible in a rack unit. With each card delivering up to 400Gbps of processing throughput, and up to four cards per server, manifold CLOUD can now achieve an unbelievable 4.8 Tbps of media processing per server. This translates to a dramatic increase in production capabilities far above any of the competing products. For Tier-One live sports producers, every camera, replay, and graphics feed from a major event like the World Cup final or Sunday Night Football can be handled on a minimal footprint without sprawling equipment racks or associated wiring, energy and cooling costs. Outside of the realm of broadcast, global E-sports productions are able to scale to hundreds of sources and outputs with focused hardware efficiency, whilst major enterprise media facilities can consolidate space, cabling, and energy consumption, whilst ensuring that their operations are future-proofed for subsequent business expansion.
Not every user will need this level of density today. But the capability underscores manifold’s position at the forefront of technological innovation in software-defined media processing – constantly pushing forward on a core idea of ‘doing more with less’. Most importantly, it demonstrates the linear flexibility of scaling with manifold: Tier One providers can deploy high-capacity 400GbE cards for maximum density, while smaller productions can start with lower-cost COTS cards and add incrementally as needs grow – all within the same unified platform, with no rearchitecting required.
Aliaksei Shauchenka, CEO and co-founder of manifold technologies, commented: “The ability to support 400GbE COTS FPGA cards is a great headline grabber, but it isn’t simply about chasing a specification – it’s about removing the barriers to meaningful creativity. Live production shouldn’t be limited by how many cards you can fit in a rack or how much power you can draw. With 4.8 Tbps per rack unit, we’re giving producers the headroom to imagine bigger while keeping their infrastructure lean”.
He added: “What excites me most is that this isn’t just for the largest broadcasters – manifold CLOUD represents a real democratisation of production capabilities. Because manifold runs on standard COTS hardware, someone can start with one card and scale to 400GbE density when they’re ready. The platform grows with you – that’s what software-defined should mean.”
More information about manifold technologies and their solutions is available at www.manifoldtech.tv.