Real multiplexes, without a national operator's budget.
Up to 100 services across 16 DVB-C frequencies from one appliance. Tables generated, PIDs discovered by probing the source, occupancy shown live, and a source swap that leaves the carrier up.
The problem
The tooling is built for someone a hundred times your size.
- Subscribers lose their channel numbers every time the line-up changes, and the numbers are what they actually remember.
- Multiplex capacity is planned on paper, and one variable-bitrate HD channel ruins the sums.
- Existing headends are priced, racked and licensed for national operators.
- A rescan across the network generates a week of support calls for a change that took ten seconds.
What Castmux does about it
A muxer, a modulator and a monitor in one process.
Real multiplexes
PAT, PMT, SDT and NIT are generated for every multiplex. Elementary stream PIDs are discovered by probing the source, not guessed from a template, so a source that changes its layout does not silently produce a service with no audio.
Capacity you can see
Live occupancy per multiplex, measured off the wire, with a warning above 90%. Capacity planning stops being a spreadsheet and becomes a number you can look at while you drag a channel into a carrier.
Hot-swap a source
Change the source behind a service without dropping the carrier. Subscribers on other services in that multiplex see nothing, and nobody has to rescan.
Drag and drop assignment
Move services between 16 frequencies in the browser. Rebalancing a multiplex that has grown too full is a drag, not a configuration file and a restart.
Stable service identity
roadmap Pinning a service id and guarding it against change, so editing the line-up cannot renumber a service underneath receivers that have already stored it.
LCN and NIT cross-linking
roadmap LCN gives every service a fixed number, so subscribers keep their positions through a line-up change. NIT cross-linking lets a receiver find every other multiplex from the first one it locks, which is what makes a network-wide fast scan possible instead of a full band sweep.
A typical installation
One box covers most small networks.
| Network size | Channels | Recommended hardware | Output |
|---|---|---|---|
| Up to 100 channels | Up to 100 | One Cable TV / IPTV Pro Box — €1,899 | DVB-C across 16 frequencies on a TBS modulator, plus UDP multicast for anything fed over IP. |
| Larger networks | More than 100 | Multiple appliances | Speak to us. Fleet management across appliances is roadmap; today each is managed on its own page. |
Net bitrate = symbol rate × log2(QAM) × 188/204. The 188/204 term is the Reed–Solomon overhead, and it is not optional. Plan to about 80% of net capacity: variable-bitrate HD services peak well above their nominal rate, and a multiplex filled to its arithmetic limit will drop packets on the peaks rather than on average. Castmux shows live occupancy per multiplex so you can see the peaks rather than assume them.
What you get
Enough instrumentation to answer a subscriber.
- Per-channel wire bitrate — measured by packet capture, not reported by an encoder
- Duplicate-sender detection — two processes on one group is named, not left as a mystery fault
- Dead-multicast-route detection — the appliance says when the network is at fault instead of restarting forever
- Per-channel logs and browser preview — confirm a service without a field visit
- All audio tracks with ISO-639 tags — on the pass-through engine, original PIDs and language descriptors survive
- OpenTelemetry metrics and an HTTP API — into your existing NOC and provisioning
- One price per appliance — not per channel, per multiplex or per subscriber
Our sister project tvheadend.io carries over 2,000 licensed television and radio services over HLS and SRT, which are the two source types Castmux ingests natively. Take a channel package and the Castmux licence costs nothing.See what is available or read thepricing. Rights differ by property type and by territory, and redistribution over a cable network is licensed differently again, so tell us what the network is and we will quote against it.
Honest limits
What Castmux does not do for a cable network.
This is the section that decides whether to book the call.
- No conditional access. No Simulcrypt, no ECMG or EMMG interface, no CA descriptors, no key management. Everything Castmux modulates goes out in the clear.
- No descrambling. No DVB-CI, no CAM slots. An encrypted feed has to be in the clear before it reaches the appliance.
- No DVB-T or DVB-S modulation. The RF output is DVB-C.
- No subscriber management, billing or provisioning. No CRM, no entitlement system.
- No must-carry or regulatory compliance tooling, and no logging built for a regulator.
- No transcoding and no statistical multiplexing. Services are carried at the rate the source produces, which is why the 80% planning figure matters.
- BISS scrambling is roadmap and is not a substitute for conditional access.
If your network needs Simulcrypt and an ECMG, this is not the product and we will tell you so on the first call rather than the fourth. Castmux is for clear networks: community cable, campuses, holiday parks, municipal and building-scale distribution.
Other solutions
Same appliance, different building.
See it running on your own feeds.
Twenty minutes, screen-shared. Bring one stream URL and we will put it on a multicast group while you watch. If it is not right for you, we will say so on the call.