Every year I go to a trade show and hear the same thing: SDI is dead, IP is the future, and anyone still running coax is holding back progress. I’ve been hearing this for the better part of a decade now.
And yet, when I’m on-site at a major live event - wiring up comms systems, routing audio under pressure, troubleshooting at midnight before a show goes on air - SDI is still there, doing exactly what it’s always done. Quietly and reliably.
I work primarily in audio and communications, so I’m not going to pretend I have no horse in this race. But I’ve spent enough time on OB trucks, in temporary broadcast centres, and at events where failure simply isn’t an option to have a fairly clear view of when each technology earns its place.
Why SDI keeps showing up
The thing about SDI is that it doesn’t ask much of you. Plug it in, it works. There’s no multicast routing to configure, no PTP timing to synchronise, no VLAN to segment. When you’re building a temporary system under time pressure with a team that hasn’t worked together before, that simplicity has real value.
It’s also deterministic in a way that IP still isn’t by default. One cable, one signal, one path. Either it works or it doesn’t - and if it doesn’t, you know exactly where to look. In a live environment, that kind of immediate fault isolation is worth a lot.
The WBD example
When it came out that Warner Bros. Discovery were building a predominantly SDI-based system for the Milano-Cortina Winter Olympics, some people in the industry seemed surprised. I wasn’t.
For an operation at that scale - hundreds of feeds, complex routing, a large team working under enormous pressure - SDI makes sense. Everyone on that crew knows it. The workflows are proven. The failure modes are understood. Introducing a large-scale IP infrastructure into a temporary setup of that complexity would add risk without necessarily adding much practical benefit.
That’s not a criticism of IP. It’s just an honest assessment of requirements.
Where IP genuinely wins
I’m not anti-IP. I think it’s clearly where the industry is heading, and for good reason. Remote production over IP has genuinely changed what’s possible - I’ve seen facilities route audio across continents in ways that would have been unthinkable a few years ago.
Where IP comes into its own is in flexibility and scale. Reconfiguring a large SDI router is a physical job. Doing the same thing on an IP network is a software change. For productions that need to adapt quickly, or that span multiple sites, that matters enormously.
The honest answer
The honest answer is that most real-world broadcast environments aren’t all-IP or all-SDI - they’re hybrid, and probably will be for some time. The instinct to frame this as a debate misses the point. They solve different problems well.
What I do think is worth pushing back on is the idea that staying with SDI represents some kind of failure to modernise. Sometimes it’s just the right tool for the job.