Are CCTV Games Live?

Are CCTV Games Live? Real Footage, Real Time, Real Outcomes

The short answer is: yes and no - and the distinction matters. CCTV games from 155.io use genuine footage captured from real-world and purpose-built camera installations. The footage is authentic. What varies by game is whether that feed is transmitted and processed in strict real-time, near-real-time, or across a controlled environment that runs continuously regardless of when you log in. Understanding this distinction is essential to understanding why the outcomes in these games cannot be manipulated.

Live vs. Pre-Recorded - A Nuanced Answer

Traditional online casino games fall into two broad categories: those driven by random number generators (RNGs), where software produces outcomes, and live dealer games, where a human dealer operates in a studio. CCTV games occupy a third category entirely.

The footage powering Rush Hour, Duck River, and Snow Run is not pre-recorded in the sense that it is archived and replayed. No round is ever the same because real physical events - traffic flow, rubber ducks moving through water, skiers descending a slope - are captured as they occur. The feed is live in the sense that it reflects what is happening right now at that location. The game mechanics, however, introduce a small layer of processing latency before that footage reaches your screen.

This is a meaningful distinction from "pre-recorded" in the way a slot game might replay a stored video sequence. The source material is real, unscripted, and unrepeatable.

What "Live" Means for Each Game

Rush Hour - Real-Time Traffic at Urban Intersections

Rush Hour draws its footage from traffic cameras positioned at real urban intersections across more than a dozen locations worldwide. Cities currently included span Tokyo, New York, London, Bangkok, Paris, Sydney, Swindon, Arizona, Taipei, Patong Beach, and Abbey Road, among others. These are functioning public infrastructure cameras or licensed equivalents - not studio reproductions.

The distinction to apply here is between real-time and near-real-time. The camera captures traffic as it passes through the detection zone. That footage is fed into 155.io's AI object detection pipeline, which processes each frame to identify, classify, and count vehicles. The AI draws bounding boxes around detected objects and updates the running count in real time. A small but unavoidable processing and encoding delay - typically a few seconds - means what you see on screen is near-real-time rather than an instantaneous mirror of the street.

Crucially, this delay is consistent across all players viewing the same round. Nobody gains an advantage from the latency because it is uniform. The traffic itself is uncontrolled: no operator or game provider has any ability to influence how many vehicles pass through a Bangkok intersection at 3am or a New York crossroads at noon.

Duck River - Continuous Operation in a Controlled Environment

Duck River is the most unusual case when discussing "live" footage. The game is set in a purpose-built 10,000 square foot warehouse facility housing a 30-metre lazy river circuit. Rubber ducks float through the course continuously, encountering obstacles that affect their trajectory and timing. The installation runs at all times - it does not stop between player sessions or overnight.

This means Duck River is perpetually live. When you log in at any hour, you are watching a real-time feed of an actual physical environment. The warehouse and its river are not shared public infrastructure - they exist solely to power this game. That gives 155.io precise control over the environment's consistency while still ensuring outcomes are determined by genuine physical events: water flow, duck buoyancy, and obstacle interaction.

The "live" question here is answered by the continuous nature of the installation. There is no pre-recorded loop. The feed reflects what is happening in that warehouse at the moment you watch it. Currently, Duck River is available on Roobet and Stake.

Snow Run - Seasonal Considerations and Multi-Angle Feeds

Snow Run, launching 7 April 2026, introduces the first outdoor alpine environment to the CCTV games portfolio. Cameras are positioned across an actual ski slope with multiple angle options: point-of-view (POV) cameras, drone footage, and slope-side static cameras. The variety of angles is part of the game's visual identity.

The seasonal dimension adds complexity to the "live" question. Snow Run is tied to an environment that exists in winter conditions. The footage is genuine alpine terrain with real skiers and snowboarders in motion. The game is confirmed on Roobet, Stake, and Shuffle for launch.

From a liveness standpoint, Snow Run operates similarly to Rush Hour: the camera captures real events in near-real-time, with AI processing introducing a brief but consistent delay before the feed reaches players. The multi-angle format means the system is simultaneously managing several camera feeds, all synchronised so players across different operators see the same footage at the same moment.

How Camera Feeds Work Technically

Understanding the technical pipeline clarifies why CCTV game outcomes are genuine. The process follows this general sequence:

1. Camera hardware at the source location (intersection, warehouse, or alpine slope) captures continuous video footage.
2. The raw feed is transmitted to 155.io's processing infrastructure.
3. AI object detection models analyse each frame, applying bounding boxes to identified objects (vehicles, ducks, or skiers depending on the game) and updating detection counts.
4. The processed feed - footage plus AI overlays - is encoded for streaming distribution.
5. The encoded stream is handed off to Hub88, the B2B aggregation platform that connects 155.io's games to licensed operators.
6. Operators (Roobet, Stake, Shuffle) receive the stream through their Hub88 integration and present it to players via their platforms.

The AI's role is detection and counting, not outcome determination. The outcome emerges from whatever happens in the physical environment. The AI simply reads and reports it.

For more detail on how the detection mechanics work, see how CCTV games work.

The Role of Hub88 in Feed Distribution

Hub88 is the aggregation layer that sits between 155.io and the operators offering CCTV games. It handles game content distribution, integration standardisation, and regulatory compliance routing across multiple jurisdictions. For players, this infrastructure layer is largely invisible - but it is why the same game appears consistently across Roobet, Stake, and Shuffle without each operator needing a direct technical integration with 155.io.

Hub88's involvement also means that operators receive the processed game feed through a vetted, auditable channel. This supports the integrity claims of the product: the feed distribution pathway is commercially and technically standardised, reducing the opportunity for tampering at any point in the chain.

Latency and Synchronisation Across Operators

One question that arises from a multi-operator distribution model is whether players on different platforms see the same footage at the same time. The answer is yes - synchronisation is a core requirement of the system.

All players watching a given round of Rush Hour, regardless of whether they are on Roobet or Stake, are watching the same footage with the same AI overlays at the same moment. The betting windows, count displays, and settlement timings are consistent across platforms. This synchronisation is maintained through the Hub88 distribution layer, which ensures the stream is delivered with equivalent latency to all connected operators.

A player on one platform cannot gain a timing advantage over a player on another. The round begins, runs, and settles at the same point for everyone.

"Real" Matters More Than "Live" for Outcome Integrity

The more important question than strict real-time delivery is whether the footage and outcomes are genuine. For CCTV games, that question has a clear answer: no party - not 155.io, not Hub88, not the operators - has any ability to control how many cars pass through a Tokyo intersection, how fast a rubber duck moves through a warehouse river, or how a skier navigates an alpine course.

This is the foundational integrity claim. Outcome authenticity does not depend on millisecond-accurate live delivery. It depends on the source of the outcome being beyond anyone's control. Real-world physical events satisfy that condition regardless of whether a few seconds of processing latency exist between the event occurring and the player seeing it.

Compare this to a traditional RNG game, where software generates outcomes. The RNG may be certified and audited, but it remains a computational process. CCTV games replace that computation with physical reality.

Comparison to Live Dealer Games

In a live dealer game, a human dealer shuffles cards or spins a wheel. That human is a point of potential error or influence. In a CCTV game, the equivalent role is played by traffic flow, water physics, or alpine terrain - none of which can be briefed, pressured, or paid.

Verification - AI Bounding Boxes, Confidence Scores, and Running Count

One of the distinctive features of CCTV games is the visible verification layer embedded in the gameplay itself. While watching a round, players can observe:

• Bounding boxes - coloured rectangles drawn around each detected object in real time, confirming what the AI has identified and where the detection zone boundaries are.
• Confidence scores - numerical indicators showing how certain the AI model is about each detection. Objects the model identifies with high confidence are counted; ambiguous detections may be flagged or excluded based on the game's rules.
• Running count - a live tally of detections within the current counting window, visible to all players simultaneously.

This transparency layer means players do not have to take the operator's word that 14 vehicles passed through a given zone during a round. The count is assembled in front of them, detection by detection, with the AI's working shown on screen. It is an unusual level of verifiability for a betting product.

Practical Next Steps

• Read the individual game guides for deeper coverage of each camera setup: Rush Hour, Duck River, Snow Run.
• For a full technical explanation of how AI detection determines outcomes, visit how CCTV games work.
• To start playing and observe the live feeds for yourself, check which operators carry each game and set up an account on one that suits your region.
• Watch at least one full round without placing a bet first - the AI bounding boxes and running count make the most sense once you have seen a complete detection window in action.