Abstract---We present a game benchmark for testing human-swarm control algorithms and interfaces in real-time, high-cadence scenarios. Our benchmark uses a swarm vs. swarm game in a virtual ROS environment to assess these algorithms and interfaces through a human operator's performance in the swarm game, which requires rapid adaptations to changes in the game state. The goal of the game is to ``capture'' all agents from the opposing swarm; the game's high-cadence is a result of the capture rules, which cause agent team sizes to fluctuate rapidly. These rules require operators to consider both the behavior of the opponent swarm and the number of agents currently at their disposal when planning any swarm actions. We demonstrate the game benchmark through our human-swarm control system (developed in previous work) which allows an operator to interact with their swarm through a high-level touchscreen interface that sends control commands to the swarm through a low-level decentralized ergodic coverage framework. We demonstrate that our human-swarm control system has traits, such as scale and permutation invariance, that are crucial for real-time human-swarm control systems operating in high-cadence environments. To encourage others to use our game benchmark to test real-time human-swarm algorithms and interfaces, we will make our code available on an open repository on Github.