Low Latency
Low latency is what makes the meeting feel conversational instead of transactional. It keeps participants from talking over each other, waiting after every point, or treating the call like a broadcast.
Use this page when you need to understand how routing, network path quality, and local setup affect the feel of the conversation.
Wireframes
Section titled “Wireframes”┌──────────────────────────────────────────────────────────────────────┐
│ Hyper app.hyper.video/call-health │
├──────────────────────────────────────────────────────────────────────┤
│ │
│ Network health │
│ │
│ Meeting: Launch readiness sync │
│ Connection: stable now, 1 short drop in last 10 min │
│ │
│ ┌──────────────────────────────────────────────────────────────┐ │
│ │ Latency 42 ms Packet loss 0.6% │ │
│ │ Jitter 6 ms Share quality Good │ │
│ │ Mic path Clean Speaker path Clean │ │
│ └──────────────────────────────────────────────────────────────┘ │
│ │
│ Suggestions │
│ - Keep camera on current quality preset │
│ - Prefer wired audio before turning on live translation │
│ - Rejoin only if packet loss rises above 3% │
│ │
└──────────────────────────────────────────────────────────────────────┘
Latency is easiest to interpret when viewed alongside packet loss, reconnect events, and current mitigation state.
When to use this
- Fast-paced discussion, workshops, sales calls, support calls, and any meeting where interruption and timing matter.
- Rollout planning for distributed teams where cross-region performance is a concern.
Before you start
- Test with the same headset, camera, and network users rely on every day.
- Watch the network panel instead of relying only on subjective call feedback.
What this feature helps with
Global edge network
Calls are routed through the nearest edge server. Wherever you are, latency stays below 100ms for natural conversation flow.
Adaptive routing
If network conditions change, calls are automatically re-routed through the fastest path — no manual intervention required.
Real-time metrics
See your connection quality in real time during calls — latency, jitter, and packet loss are all visible so you know what's happening.
How to work with it
Join a meeting
- Hyper automatically connects to the nearest edge server
- Optimal routing is determined in milliseconds
Enjoy natural conversation
- Sub-100ms latency feels like being in the same room
- No delays, no echo, no talking over each other
Network adapts automatically
- If conditions change, routing adjusts in real time
- You stay connected without interruption
Where it fits well
Global teams
- Cross-continent calls with minimal delay
- Edge servers on every major continent
- Consistent quality regardless of distance
- Natural conversation flow across time zones
Real-time collaboration
- Pair programming sessions without lag
- Design reviews with instant feedback
- Whiteboarding with real-time updates
- Music and audio-sensitive work
Sales calls
- No awkward pauses during pitches
- Natural rapport-building with prospects
- Screen sharing without delay
- Professional experience for clients
What to expect
- Sub-100ms latency
- Global edge network
- Adaptive routing
- Real-time quality metrics
- Automatic failover
- No configuration needed
Questions that come up often
What is typical latency?
Most calls experience 30-80ms latency depending on geographic distance. Our global edge network ensures it stays below 100ms for the vast majority of connections.
How does adaptive routing work?
Hyper continuously monitors network conditions and re-routes calls through the fastest available path if latency increases or packet loss is detected.
Can I see my connection quality?
Yes. Real-time connection metrics are available during any call. See latency, jitter, and packet loss in the call info panel.
Related tasks
Limitations and rollout notes
- Local device load, VPN routing, and restrictive networking can still increase end-to-end delay.
- Cross-region geography always matters, even with edge acceleration and adaptive routing.
Troubleshooting
People keep speaking over each other
- Check if latency is consistently elevated rather than only briefly spiking.
- Compare affected users against the rest of the meeting to isolate whether the problem is global or local.
- Review whether local CPU pressure, Bluetooth routing, or VPN traffic is adding delay before blaming the meeting service.