One of the use case we have is to have max 2 video presenter with 1 screensharing simulteneously and bunch of people watching, listening and possibly asking questions. Let us say 80%-20% p2p and hub-spoke distribution.
- For the best audio quality how many people can attend? Is 30 too much?
- Has this been done before? What were the challenges?
The issue here is that we can not really test this scenario(unless you have ideas) since gettting 30 people is hard. Possibly we can check for data loss but even that is not straigtforward. So the first real test will be in front of the client. If this use case is not likely work with good audio quality we can always suggest using conference lines for audio portion. We are planning to implement the dynamically changing audio quality with bandwidth.
Any first hand experience with this use case? Any live example where someone is using this?
Yes its been done before. We did have customers with this scenario and we did optimize our AudioPublishers & AudioSubscribers to handle scenarios of 30+ users in a session.
I wish I could promise you quality for 30+ users and number of users for optimal audio quality, but then it depends on various other factors.