The Video Quality Widget test the video throughput in WebRTC when connecting to a media server. The media server being connected to depends on the infrastructure used. This can be used to understand the connectivity and quality that can be achieved in calls requiring connecting to media servers (such as group video or recorded sessions).
This tests connects a WebRTC peer connection directly to the media server of the tested infrastructure, sending and then receiving back the same video through the media server and checking the incoming bitrate achieved.
This gives a good approximation of what can be expected in a video session through the media server. It should be taken into account here that:
- The estimates are based on the browser and are done over a short period of time – available bandwidth changes dynamically
- The video received is an echo of what is being sent to the media server. This means it is capped at the uplink bandwidth available and is based on the complexity of the video being sent out (a camera privacy cover will reduce the video bitrate being sent and by extension, the video bitrate being received)
Data we collect and share
The columns indicate incoming and outgoing media related metrics. As networks are asymmetric in nature, they offer different insights towards your connection quality.
| Estimated | The bitrate that the WebRTC library estimates it can send over the network without interference. This is provided only for the outgoing media stream. |
| Bitrate | The effective video bitrate during the test. |
| Round Trip Time | The round trip time for media, as calculated on the incoming stream. |
| Packet Loss | Packet loss percentage observed during the test. |
Things to notice
If you are using a camera cap, please make sure to have it open for this test.
You want packet loss and round trip time to have a low value.
The bitrate should be high enough to the expected call quality you are after.
If the bitrate value is below 30kbps and then estimated bitrate is high, it might be due to the fact that the camera used has static input to it – usually because of a closed camera cap. Be sure to have live and dynamic content for the camera for this test.