Walk into the production compound for a College Football Playoff championship game, and you’ll find a small city of mobile units, hundreds of crew members, and over a dozen distinct broadcasts being produced simultaneously from a shared IP backbone. Each feed has its own editorial voice, its own talent, and its own production chain. It’s one of the most complex live production operations on the planet.
That same week, a university athletic department across the country produces an ESPN+ basketball game with four cameras, two students, and a single switcher in a campus control room.
Both are live sports broadcasts. Both need to work flawlessly. And both run on production infrastructure that operates the same way it has for decades. Switching, routing, and processing capacity is fixed to a specific truck or control room, unable to flex, share, or scale when the next production on the schedule needs something different.
That model held up when the production calendar was predictable and the output was a single linear feed. It’s breaking under the weight of what sports broadcasting has become.
Simultaneity, not complexity, is the real infrastructure problem
The challenge isn’t any single production. A national primetime NFL broadcast is complex, but the industry has decades of experience producing them. The challenge is how many productions need to happen at the same time, on the same day, drawing from the same finite pool of equipment and crew.
On peak nights during NBA, NHL, and MLB season overlap, major OB companies deploy every truck and every crew they have. The constraint isn’t talent or ambition, but rather raw switching and processing resources. Each independent production chain needs its own set of M/E banks, its own replay servers, and its own graphics channels. Fifteen games on the same Tuesday night means fifteen fully independent production environments, regardless of whether each one actually needs the full capacity it’s been allocated.
REMI was supposed to relieve this pressure. Moving switching to centralized hubs instead of deploying trucks to every venue does reduce the logistics of getting equipment in place. ESPN now runs more than 3,500 annual REMI event hours across seven dedicated facilities. But centralization doesn’t eliminate the bottleneck. It just moves it somewhere else.
A centralized production hub still needs discrete switchers, replay servers, and routing for each control room. When five games happen simultaneously, the hub needs five independently equipped production environments, just like the old model needed five trucks.
More games, more platforms, and more feeds per event are multiplying demand
The production volume the industry carries keeps growing, and the forces behind it are permanent.
Rights fragmentation is the most visible driver. Leagues are splitting their packages across more partners than ever, and each rights holder requires its own branded production with unique studios, talent, graphics, and camera philosophy. The NBA’s latest deal, for example, splits games across three national partners instead of two, plus league-owned channels and direct-to-consumer platforms.
More rights holders means more separate production chains competing for the same finite infrastructure on the same game nights.
Output multiplication within a single event compounds the problem. The alternate broadcast is no longer a novelty layered on top of the main telecast. It’s a core design driver for the production compound itself.
A single Thursday Night Football broadcast generates well over 100 distinct feeds across its broadcast streams, each with redundant transmission paths. Coach’s film rooms, betting data feeds, Spanish-language broadcasts, vertical mobile-first formats, social clip packages—every alternate output requires its own switching resources that a traditional two to three M/E production switcher was never designed to provide.
At the collegiate level, the volume is staggering. ESPN presented more than 30,000 live events in the 2024-25 academic year. Schools are producing much of this content themselves with minimal infrastructure—small full-time staffs supplemented by freelancers and students, running ESPN+ broadcasts out of campus control rooms never built for this kind of output. The production is getting done. But the infrastructure underneath it is stretched to its limits.
Live sports production is getting bigger, smarter, and closer to the fan
Watch this TV Tech webinar on demand as panelists from FOX Sports, Ross Video, AJA, and Appear explore how the sports broadcasting industry can cover more events, deliver content to more platforms, and tell better stories without simply adding more people, trucks, or time.
The RSN collapse is forcing a production infrastructure reckoning
The financial foundation that sustained local sports broadcasting for three decades has effectively collapsed. The regional sports network model—built on guaranteed carriage fees from cable bundles—is unwinding as cord-cutting accelerates. FanDuel Sports Network confirmed its networks would close by mid-April 2026. MLB Local Media now produces and distributes games for 14 of 30 MLB teams, with additional teams running independent operations.
Teams moving to these new distribution models are seeing roughly half the revenue they earned under their old RSN deals, with payments now tied to streaming and advertising performance rather than guaranteed rights fees.
Only 30% of Americans still use traditional pay-TV. This revenue compression puts direct pressure on production infrastructure investment. Teams and their broadcast partners need professional quality on dramatically smaller budgets, which means the infrastructure they invest in has to work harder across more situations than ever before.
Fixed production infrastructure sized for peak demand is the industry’s hidden cost problem
Every sports broadcaster, OB company, and athletic conference faces the same infrastructure trade-off. A production switcher in a truck runs at full capacity on game day, then sits idle the rest of the week and goes dark during the off-season. A centralized hub runs five simultaneous productions on a packed Saturday, but only one or two on a weeknight. The equipment costs the same either way.
The traditional answer has been to size for peak. Equip every truck, every control room, and every production environment for the most demanding scenario it might face. A fully equipped HD mobile unit can cost millions to build, with timelines running over a year from design to deployment. University athletics departments are spending $5 to $8 million on dedicated broadcast centers just to keep pace with ESPN+ obligations.
That investment is justified by the handful of days each year when the infrastructure runs at capacity. The rest of the time, it’s idle capital that can’t be redirected, shared, or redeployed. An M/E bank sitting unused in a truck parked at the depot can’t help the production happening at the arena across town.
The infrastructure model that serves that mandate can’t be one where every production environment carries its own peak-ready capacity independently, with no way to share what’s not being used.
Frame the game your way
Built for the pressure of live sports, Ross brings cameras, replay, graphics, and automation together in a powerful system that helps you capture more angles, add more insight, and share more live data to keep your audience hooked.
Software-defined production infrastructure is moving from concept to deployment
The alternative is to use software-configurable production systems that allow operators to adjust license assignments and make better use of the processing capacity installed in each frame. A marquee rivalry game may require a four M/E configuration, while a mid-tier regular-season game may require two and a streaming-only broadcast one. When the rivalry game ends, those resources return to the pool and are available for the next production on the schedule.
This isn’t theoretical anymore. Real deployments are running this way.
- NEP’s Andrews Hub in Sydney handles up to six simultaneous Tier 1 sports productions connected to 29 venues over a fully IP infrastructure.
- CBS Sports’ Golazo Network runs entirely on cloud-based switching.
- Pac-12 Enterprises operates software-defined production rooms alongside traditional REMI control rooms in the same centralized facility.
- The EBU and Linux Foundation are developing open interoperability standards to make this work across vendors.
Some degree of resource flexibility within a single switcher is already established in parts of the market. Hyperconverged production platforms take that principle further by combining routing, switching, and processing in unified frames with software-licensed feature sets, allowing systems to be configured for a broader range of productions. The majority of sports OB trucks still run traditional dedicated switchers, and TD familiarity with existing systems creates real inertia. But the architectural direction is clear, and the economics of idle infrastructure are pushing the industry toward it.
Ross Video’s Carbonite HyperMax is a software-defined, hyperconverged production platform that combines production switching and other processing tools while allowing operators to configure licensed capabilities for different productions.
The infrastructure model needs to match how sports broadcasting actually works
Sports broadcasting is defined by variable demand. A packed playoff Saturday looks nothing like a quiet Tuesday in the off-season. A 14-feed MegaCast compound looks nothing like a 4-camera collegiate stream. The production infrastructure serving all of it needs to support flexible configurations rather than being permanently sized for only one level of demand.
The organizations that figure out how to manage and configure their existing production resources to match demand can operate more efficiently and produce better content. The technology to do that is no longer theoretical. The question is whether sports broadcast operations are ready to move toward it.