Simulcast
What is simulcast?
Simulcast is a video technique in which a sender encodes the same camera picture at several qualities at once, so a media server can pass each viewer the version that fits their screen and connection. Nobody has to agree on one quality for the whole call: a viewer on a weak connection or with a small tile gets a lighter stream, while everyone else keeps a sharp picture.
How does simulcast work?
With simulcast turned on, the sender's browser encodes the camera picture more than once at the same moment: for example a small low-resolution version, a medium one and a full-quality one. All of them go up to the media server, an SFU. The server then forwards exactly one version to each viewer.
The choice is made per viewer and changes during the call. The server looks at how large the tile is on that viewer's screen and how much bandwidth their connection has right now. If the viewer enlarges a speaker or their network improves, the server switches them to a better version; if their connection drops, it switches them down, without asking the sender to change anything.
| Viewer | Typical version |
|---|---|
| Small tile in a large grid | A low-resolution version; a thumbnail cannot show more detail anyway. |
| Active speaker on a large screen | The full-quality version, if the connection can carry it. |
| Phone on a weak mobile network | A lighter version, so the picture keeps moving instead of freezing. |
Why does simulcast matter in a video call?
Without simulcast, a server can forward only the one version the sender produces. Either the sender lowers quality for everyone so the weakest connection can follow, or the weakest participant receives more than their network can carry and sees frozen video. Simulcast removes that trade-off: each person gets the best picture their own screen and connection can use.
The cost sits with the sender, whose device encodes several versions and uploads a little more. That is why calling software reduces the number or size of versions on slower computers, and why a weak upload connection still lowers what everyone receives from that person.
How does Meeto use simulcast?
In Meeto's web app, camera video is sent to the media server as several versions, and each viewer receives only the quality their tiles need. The top version depends on the camera, the browser and the computer, so not every device sends the same set.
When a connection stays poor, the call steps down in stages: it lowers the camera's bitrate, then shows only the active speaker's video with the notice “Weak network — video only from the speaker” and a “Show everyone” button, and finally switches to audio only. It returns to full video by itself once the network recovers. The laggy call guide covers what you can check on your side, and the TURN server entry explains how a call gets through strict networks.
Frequently asked questions
Does simulcast use more bandwidth?
For the sender, a little more: the upload carries the extra smaller versions alongside the full one, and the small versions are cheap compared with the top quality. For viewers it usually means less, because each one receives a single version that fits their screen instead of the largest one.
What is the difference between simulcast and SVC?
Simulcast sends several independent encodings of the same picture. Scalable video coding (SVC) sends one stream built in nested layers, so a server can drop the upper layers for a viewer. Both let a media server serve different qualities; which one a service uses depends on the codecs and browsers it supports.
Can I choose the video quality myself in Meeto?
No. Meeto has no manual video quality setting; the quality each viewer receives is chosen automatically. The layout you pick in View, such as Grid or Speaker, changes which tiles are large on your screen, and larger tiles get a sharper picture.
Try your first browser meeting
Create a room, share the link and invite someone to join you.