Most people picture broadcast television as finished technology, more or less frozen in time. It isn't. Brazil is designing the next generation of its terrestrial TV system right now, and the interesting part for engineers is how it's being put together: an IP-centric platform that borrows the best pieces from several different standards.
I work on TV 3.0, so this is the introduction I wish I'd had the first time someone asked me, "wait, isn't TV just... TV?"
What we have today
Brazil's current digital TV system is SBTVD (also called ISDB-Tb), an evolution of Japan's ISDB-T. Two things about it matter for what comes next.
Transmission runs on a dedicated broadcast physical layer: robust, and designed for antennas and for reception on the move. Interactivity runs on Ginga, the middleware shipped in millions of TVs and set-top boxes across Brazil and Latin America. Ginga is what lets a broadcaster deliver an app alongside the video.
It works well, but it was designed before streaming, 4K, HDR and second-screen everything became the baseline expectation. That gap is what TV 3.0 closes.
What TV 3.0 is
TV 3.0 is the next-generation terrestrial TV system defined by the SBTVD Forum through an open call for proposals. Instead of adopting one foreign standard wholesale, the Forum evaluated each component on its own and assembled the stack piece by piece. The result is unusual, and to me pragmatic:
| Layer | TV 3.0 choice |
|---|---|
| Physical (over-the-air) | ATSC 3.0 physical layer |
| Transport | IP-based (ROUTE/DASH) |
| Video coding | VVC (H.266) |
| Audio | MPEG-H 3D Audio |
| Application / interactivity | Ginga, evolved |
The part that tends to surprise people: Brazil kept Ginga for the application layer, but adopted the ATSC 3.0 physical layer (from the US system) for transmission, and put IP at the center of everything.
The core idea: it all becomes IP
This is what matters most if you write software. In TV 3.0 the transport is essentially IP delivered over the air. Broadcast stops being a closed, opaque pipe and starts behaving like the internet.
In practice:
- Content travels as IP using ROUTE/DASH, the same DASH you already know from streaming.
- Because everything is IP, a receiver blends broadcast and broadband seamlessly. The 4K feed can arrive over the air while personalized or on-demand content comes over your home connection, and to the viewer it's a single experience.
- Modern codecs (VVC for video, MPEG-H for audio) let 4K/8K, HDR and immersive sound fit in the available spectrum.
Broadcast and the web stop being two separate worlds. They meet in the middle.
Why it matters
Sharper picture and sound are the easy pitch: 4K, HDR, immersive audio. But the architecture opens up more than that.
You get hybrid experiences, where live TV is enriched with synchronized content pulled from the internet. Interactivity becomes central instead of an add-on: with Ginga and web technologies, a broadcaster can ship an app next to the signal, from accessibility features to extra camera angles to real-time information. And none of this gives up reach. Free-to-air TV still gets to homes that streaming alone doesn't, and TV 3.0 modernizes that reach rather than replacing it.
For a country the size of Brazil, bringing internet-grade features to free antenna TV is a big deal.
Why it's interesting to build
From an engineering standpoint, TV 3.0 sits at a rare crossroads: embedded systems (it runs on constrained TV and set-top-box hardware), multimedia (codecs, streaming, synchronization) and web development (the interactive layer). You can spend the morning in low-level C/C++ and Linux and the afternoon reasoning about HTML, DASH and app behavior. Few platforms ask for that range.
What's next
TV 3.0 is moving out of specification and field trials toward rollout. When it arrives, the TV in the living room will be running an IP stack, modern codecs and an app runtime, much closer to a web platform than to the analog set it descends from.
Top comments (0)