This study investigates the communication performance of wireless systems emulating Virtual Reality (VR) workloads in industrial computing environments. Industrial VR applications increasingly rely on distributed architectures, where VR devices exchange high-frequency data with industrial computers and cloud or edge servers, placing strict requirements on latency and jitter. To analyze these constraints, a controlled experimental testbed was established using an industrial-grade computer, a dedicated server, and multiple parallel client processes transmitting VR-like tracking and interaction messages over a wireless network. A custom ZeroMQ-based measurement framework was developed to generate deterministic, high-frequency bidirectional traffic representative of real VR communication loops. Each experimental scenario produced more than 40,000 round-trip time (RTT) measurements across parallel message streams, enabling statistically robust analysis of latency, jitter, and percentile-based performance metrics. The evaluation focuses on mean latency as well as higher percentiles (P90–P99.9), which are critical for identifying worst-case delays that can disrupt VR immersion. The results show that average RTT values remain consistently low, typically around 4–5 ms, indicating that wireless links can support many industrial VR communication tasks. However, sporadic high-latency spikes were observed at upper percentiles, reflecting the non-deterministic nature of wireless networks under load. These findings provide a quantitative baseline for assessing the suitability of wireless communication in industrial VR systems and highlight the need for predictive mechanisms, edge processing, or hybrid architectures to ensure reliable performance in latency-sensitive VR applications.