News & Trends

You know the scene. You’re days from go-live. The client wants “one more feed” on “one more screen.” Your AV-over-IP build turns into a blame loop: encoder vendor blames decoder, decoder blames the switch, switch vendor blames “the network,” and you get the phone calls.

That mess has one root cause: too many boxes that claim to be “open” but do not behave the same way.

IPMX certification is meant to cut that risk. The first official IPMX Product Testing and Certification Event took place 19–23 January 2026 at EBU headquarters in Geneva and gave manufacturers a simple outcome: pass or fail. AIMS, VSF, AMWA, and the EBU are positioning it as the moment IPMX stops being a promise and starts being something you can check. The cleanest statement of the dates and intent sits in the AIMS partner announcement.

For European integrators, consultants, and enterprise AV/IT managers, that matters for one reason: it gives you language you can write into tenders and enforce at handover.

IPMX week is about proof, not claims

Pro AV has treated “standards-based” like a free pass. It isn’t. A standard only helps when products implement it properly and consistently, and when someone checks.

IPMX is trying to earn trust the hard way: by tying a Pro AV-focused spec suite to a repeatable test plan, then running products through it in a neutral lab environment. Geneva proved to be a fitting venue, as the EBU is known for hosting serious technical events. Its technology arm has already treated IPMX as a serious topic in broadcast engineering circles, including its public IPMX for Broadcasters material from its Network Technology Seminar programme.

This is the part procurement understands. “Supports IPMX” is a marketing line. “Certified” is a pass/fail result tied to published requirements.

What IPMX is built on, and why pro AV needs the extra work

IPMX sits on the same foundations broadcast engineers have been using to push media over managed IP networks. That world is strongly shaped by SMPTE ST 2110, which defines how professional media essences travel, stay synchronised, and describe themselves over IP. SMPTE’s own overview of the ST 2110 standards suite makes clear what it targets: professional media over managed IP, with strict expectations around timing and behaviour.

Most Pro AV jobs do not look like broadcast plants. You inherit mixed networks, mixed clocks, mixed endpoints, and IT rules that were not written with multicast media flows in mind. You also inherit clients who change requirements late. Those conditions punish small gaps in device behaviour. A box that “mostly” behaves can still ruin a project when it meets another vendor’s “mostly”.

That is where IPMX steps in. The IPMX effort, via the VSF TR-10 suite, spells out Pro AV-relevant requirements, including how devices should behave and how control and discovery should work. IPMX’s own site flagged this direction early when it published the VSF TR-10 draft announcement and framed TR-10 as the layer that adapts the broadcast stack for wider professional markets.

What IPMX actually standardises, in practical terms

Keep it simple. IPMX tries to make two parts of AV-over-IP less painful: media flow and control.

Media flow is video and audio moving across the network. If a sender or receiver behaves oddly, you see it as dropouts, missing streams, odd latency, or “it works until we add the next room”. IPMX aims to reduce these surprises by setting clear requirements for devices and system environments.

Control is the other pain point. If you don’t have a sane way to discover endpoints and connect flows, you end up with manual configs, fragile scripts, and support calls that never end. IPMX leans heavily on AMWA NMOS so controllers can discover what is on the network and connect it in a repeatable way. You can see the centre of that relationship inside AMWA’s NMOS specifications, which cover discovery/registration and connection management in concrete API terms.

The detail that matters for buyers is not the acronym soup. It’s the outcome. A system that uses a common control and discovery model is easier to commission, easier to document, and less likely to collapse into bespoke glue.

What certification is likely to check, and why you should care

Certification only helps if it tests the things that break in the real world.

The VSF TR-10 suite breaks IPMX requirements into pieces, and one of the most relevant for day-to-day Pro AV deployments is NMOS behaviour. VSF TR-10-8, for example, covers NMOS requirements for IPMX systems and points implementers at the NMOS API expectations that drive interoperability. That document exists in a published, dated form as VSF_TR-10-8_2024-02-23.

For you, this matters because it narrows the “mystery failure” zone. When both endpoints implement the same NMOS behaviour, you get fewer situations where one vendor discovers devices one way and another vendor expects a different data model. You still have to design and operate the system, but you spend less time fighting basic interoperability.

It also pushes the industry towards a more honest split between products that truly support interoperable control planes and those that only claim it.

Security does not disappear: certification makes it harder to ignore

AV-over-IP is not just media planes. It is control planes, and control planes live on networks with real threat models. If you deploy discovery and connection APIs without proper security, you create a routing playground for anyone on the network. That is not a theoretical risk. It’s a procurement and governance problem.

AMWA has treated this as a first-order issue and published clear best current practice on securing NMOS APIs. BCP-003 lays out practical recommendations like TLS for API communications and OAuth-based authorisation for access control. That guidance is captured on AMWA’s own BCP-003 Security recommendations.

This is the bit you should surface early with client IT. When you say “IPMX certified”, you should also say how you will secure the control plane. If you don’t, IT will do it for you, and they will do it in ways that can break commissioning if you show up unprepared.

When “plug-and-play” is real, And when it is a fairy tale

If you run certified endpoints on a network built for media traffic, IPMX can feel close to plug-and-play. Devices appear in control. You connect a source to a screen. It holds steady when you add another receiver or reboot a switch.

If you throw it on a flat office LAN and hope, it won’t. Video over IP punishes hope. It fails fast and it fails loud: dropouts, missing streams, jitter, and support calls that turn into politics.

Certification does not replace engineering. It reduces one of the biggest wastes of time on mixed-brand jobs: arguing about whose box is at fault when both boxes should have followed the same rules. If you remove that argument, you can focus on the real culprits: multicast configuration, bandwidth planning, timing strategy, and how the control layer integrates into the wider system.

Listen Now 🎙️: Building Bridges, Not Fences: How IPMX is Connecting ProAV and Broadcast 

What to change in your 2026 tenders and specs

The Geneva event has provided a clean lever for writing IPMX certification into buying documents, and you can turn it into usable tender language fast. If you want “IPMX certification” to be something procurement can actually enforce, you need two things: tender wording that doesn’t read like marketing, and an acceptance test plan that gives you a clean pass/fail at handover. The widget below generates both in seconds, plus bidder questions that force evidence instead of promises.

Pick a preset (UK public sector, German enterprise campus, Nordic control room, venue/live) or set your own project parameters, then copy the outputs straight into your tender pack.

Treat the outputs as a starting point: good but not gospel. Keep the clauses short, make “certification evidence” a submission requirement, and make interoperability a handover gate when you’re mixing brands. If your client’s network is IT-owned or shared, the generator’s “network assumptions” language is the bit that saves you from late-stage fights about multicast, QoS, and who owns what.

Start with clarity. Now that IPMX certification is an established process, ask for IPMX certified AV-over-IP where you need mixed-brand choice and where the client expects the system to survive refresh cycles. Use less strict language only where price dominates and the client accepts lock-in risk.

Then demand evidence. “Roadmap” is not evidence. “IPMX-ready” is not evidence. “Certified” is evidence. If a bid can’t show certification, treat the claim as future capability and price the risk as a change order waiting to happen.

Next, tie it to acceptance tests. If the client pays extra for cross-vendor interoperability, your handover must test cross-vendor interoperability. That sounds obvious. It still gets skipped. Your best defence is contract language that makes interoperability a pass/fail handover item. If the build can’t connect and hold stable under the agreed test conditions, you don’t sign it off.

Finally, force the AV/IT conversation early. IPMX sits right on the AV–network border. That border is where projects die. If IT tries to treat media streams like office traffic, you will learn about it at the worst possible time unless you surface it at design stage and agree rules for multicast, QoS, segmentation, and API access.

Where IPMX certification will show up first in Europe

Expect early traction where buyers already value repeatable behaviour and vendor choice.

Let’s talk geography. The first real pull will come from countries that already have ST 2110 and NMOS muscle memory in broadcast and live production, because those teams understand why “certified” beats “supports”. Switzerland is an obvious early cluster simply because the testing week sits in Geneva and the region already runs serious all-IP broadcast estates. The UK should see early interest wherever broadcast-adjacent teams and corporate media hubs already work in ST 2110/NMOS-style architectures (BBC Wales’ Cardiff build is a clean example of that approach in practice).

The Nordics also sit in the “early adopter” bucket: Norway’s NRK has already been deep in IP production transformation, which tends to spill over into venue, events, and enterprise video teams. Germany won’t lag on capability either — when ARD/ZDF training is explicitly hands-on with ST 2110 and NMOS, you can expect consultants and integrators to start treating “interop proof” as a normal tender ask. 

You will see it in broadcast-adjacent venues and corporate media hubs, because those teams already understand the logic of ST 2110-style managed IP and they already feel the cost of downtime. You will see it in high-value enterprise deployments where the AV/IT owner wants a standard estate, not a set of one-off rooms. You will also see it in premium education and public sector sites where procurement wants formal proof, not marketing assurances.

You will see slower uptake in cost-driven jobs where the buyer is happy to accept single-vendor stacks and short refresh cycles. That is fine. Not every project needs the overhead. The mistake is pretending every project can live on hope and goodwill.

My Verdict

IPMX certification is the first serious step towards open, test-backed AV-over-IP you can spec and police.

If you build systems that must last, start writing IPMX certification into 2026 tenders and make bidders show proof. Use our tender spec generator to get you heading in the right direction. Tie it to an acceptance test that runs mixed brands under real load. That saves you from late-stage interop firefights and makes your support model saner.

Don’t sell it as magic. It won’t fix weak network design. It will stop one of the worst time-wasters in Pro AV: “open” boxes that don’t act open.

If you missed the IPMX event, you can still explore the certification program and its requirements through AIMS and its partner organizations.

Main photo credit: Getty Images/Edwin Tan