RTP Packet Loss Overview
RTP packet loss shows how many media packets failed to arrive during a call, meeting, broadcast, or stream. It matters because real time audio and video cannot wait like file downloads. Missing packets create silence, robotic speech, frozen frames, or blurred motion. A small loss rate may be acceptable for resilient codecs. A high rate usually points to congestion, weak Wi-Fi, overloaded routers, or poor routing.
Why Sequence Numbers Matter
Each RTP packet carries a sequence number. The receiver can compare the first and highest sequence values to estimate how many packets were expected. Received packets are then compared with that expected count. The difference becomes lost packets. Wrap cycles are important because RTP sequence numbers are limited. Long streams can pass the maximum value and start again. This calculator lets you include those cycles.
Reading The Result
Loss percentage is the main health signal. Delivery percentage shows how much media arrived. RTCP fraction lost converts interval loss into a 0 to 255 report value. A value near zero is healthy. Larger values show stronger recent loss. Goodput estimates payload delivery rate. Lost media time converts packet gaps into seconds, using the frame time you enter.
Jitter And Late Packets
Jitter is variation in packet arrival time. A jitter buffer can hide small variation. Large jitter can still cause packets to arrive too late. Late discarded packets may sound like normal loss to the listener. This tool separates duplicates, late discards, and adjusted received packets. That makes troubleshooting easier.
Practical Network Use
Run the calculation for each direction. Uplink and downlink can differ greatly. Compare results across access networks, codecs, and time periods. If loss rises during busy hours, bandwidth or queue management may need attention. If Wi-Fi shows higher loss than wired tests, inspect signal strength, channel overlap, and roaming. Use exported CSV or PDF reports to document tests, share findings, and track improvements after changes.
Best Testing Method
Measure during normal traffic, not only during quiet periods. Keep packet capture clocks correct. Record codec, packetization, device model, and route notes. Repeat tests after every network change. Consistent inputs make trend reports reliable and help teams separate random spikes from repeatable problems with greater confidence.