Simulcast was the WebRTC industry's first answer to "how do we send one upload to viewers with widely different bandwidth?" Instead of encoding once and trying to make every viewer fit, the publisher encodes the same source at three resolutions — typically 1080p, 540p, 180p — and uploads all three layers to the SFU. The SFU then forwards only the layer each receiver can handle. A receiver on a fast connection gets 1080p; a receiver on cellular gets 180p; the publisher's upload is constant.
Simulcast costs the publisher CPU and uplink bandwidth (3× single-layer for typical 3-layer simulcast), but eliminates per-receiver re-encoding at the SFU. It is supported in all major WebRTC stacks — Chrome, Firefox, Safari, mediasoup, LiveKit, Janus — and is widely deployed in production video conferencing (Google Meet, Microsoft Teams, Zoom).
Simulcast has a successor in SVC, where the encoder produces a single scalable layered bitstream and the SFU drops layers without re-encoding. SVC is more bandwidth-efficient than simulcast (1 stream vs 3) but harder to encode and historically had patchy browser support. By 2026 SVC support via VP9 and AV1 has matured and is replacing simulcast in newer deployments, though simulcast remains dominant for H.264-based legacy stacks.
