Videoconference performance

I did not try it, does it satisfies my requirements?

jitsi meet is xmpp :slight_smile: Here is a tutorial on how to use it with an existing Ejabberd server:
The clients of course need to support COLIBRI video bridging.

Ah, and here is a high performance Jitsi-Meet instance hosted in Germany:


I discovered that jitsi has completed the code of Jitsi Video Bridge (JVB) 2.0 and it close to release it for production.
A simple solution is to add VP9 support to it. Any available programmer is welcome to participate.


I gave it up with Talk. It only works with 3-4 participants “somehow”. To get it seriously working you need the “high performance backend” starting from 4.500 € :neutral_face:


No, it need on clients side high performance machine. And maximum is 4-5 participants.

Sorry - I’ve corrected my post.

Yes, it works with 3,4,5 people. But above no way :confused:

1 Like

It’s a shame, because it’s free to use OnlyOffice in NextCloud for up to 20 users, it seems. It would be good if the Talk application could also allow 20 participants for free at a time during a video conference.

Request created for Jitsi Integration in Nextcloud Talk
See these requests for WebDAV and CalDAV support in Jitsi


Looks like JSXC should soon be updated for Nextcloud 18. By itself it also does only WebRTC video calls (like Nextcloud Talk) but it is quite likely that Colibri videobridge could be supported as that is the same tech Jitsi meet uses (both JSXC and Jitsi Meet are using XMPP under the hood).


I opened a feature request for integrating Kurento or OpenVidu into Talk here:

I found this quite recent paper of 2018 that compares: jitsi, kurento, janus and medooze. Jitsi and medooze were the best SFU in terms of number of users (490 users with 70 rooms) and RTT while medooze was the best in terms of image quality.
Another powerfull SFU that is missing from the paper is mediasoup which is the backend of multiparty-meeting.

Even if it is a side topic here: performance of SFUs is crucial. In case of Jitsi Meet, the videobridges should be load balanced and run on decent hardware. We have been experimenting a lot with virtualization and k8s containers but finally put them on bare metal, now handling thousands of conferences per day with room sizes up to 50 participants and more.

1 Like

Read for Jitsi Meet - integration in Nextcloud Talk:

Hi everyone,
I’m trying to improve the documentation and we’re also working on more settings and other optimizations to improve performance of Talk.

In theory, it should be possible to get about 20 participants in a single call without the HPB, but doing this requires a number of things. One limit is that a single video stream is about 1 mbit/sec so you can imagine that with 5 participants, each is already sending 4 streams of 1 mbit/sec and that quickly stops scaling. EDIT: Often, the problem is on the decoding side as well, decoding 5-6 video streams of 640x480 at 30 FPS is a lot of work and overloads systems. We are looking at how we can lower that, with decreasing quality and finding codecs that are commonly hardware-accelerated. Help welcome.

We definitely want to improve that, there are issues open for this but our backlog is sadly pretty big. We make progress quickly, just look at the changelogs from releases, 8.0.6 fixed some bottlenecks and we’re doing more.

However, you can do something yourself: simply not use video. That saves a lot of bandwidth, of course. You suddenly only send 50kbit/sec instead of 1 mbit/sec. If 1 participant has video or screensharing and the rest audio-only, you should be able to have that 15 to 20-ish people in a call. (EDIT: earlier I wrote you can’t disable video during a call, you will have to disable it, leave and join again, but that was actually fixed in a rewrite of the calling code earlier, we just tested it and 8.0.x should allow you to just disable video and lower the bitrate by ~20x)

I know it isn’t ideal and we will have to add the ability to simply lower the bandwidth being used (eg use 200kbit/sec video instead of 1 mbit/sec - we are working on this and might be able to bring this in the 8.0.x series still) and other features during a call but that is simply work we have not yet done and seriously, help is welcome. The talk team has loads of customers pushing the team to implement things those customers need. They pay the bill, you know, hard to ignore them.

In the mean time, obviously Jitsi is a nice solution and yes, if you can set that up, use it… There’s no need to hate on open source solutions for not doing everything you want. Pull up your sleeves and help out if you need them to do something instead.

Anyhow, I’m creating and discussing a PR here to the that gives some tips on scaling up, feedback is welcome:

If you want to help, here a few issues that will make a big difference:


@jospoortvliet this is fantastic. Thanks for jumping in and sharing the technical details. This is exactly the kind of information I love being able to quickly find and understand in regards to Nextcloud. Cheers!

Sir @jospoortvliet Thank you for the information and looking forward for NC Talk enhancement and to use without HPB to get about 20 participants.

as of now we setup our own server for video conference using jitsi and very nice videocon open source solution

uisng only 4core cpu and 5gb memory Ubuntu 18.04 had nice audio and video quality with 10 to 30 participants.

With this looking forward also to NC can integrate Jitsi

Thank you


The only real way to compete against Jitsi and BigBlueButton in the selfhosted Open Source space is an Open Source SFU/MCU like Kurento or OpenVidu that works with Talk. Not even Geant advises Talk but rather Jitsi as a conference tool. I think you have to update your strategy a bit.

1 Like

We just don’t have the resources to build a back-end, sorry. Also, we don’t need to replace every other solution in the world, BigBlueButton and Jitsi are great as they are and people should use them when appropriate. Talk will improve but we will focus on the end user experience and leave the underlying tech to others, like we do with everything (we don’t develop file systems, LDAP, mail servers etc etc either).

And as I’ve said before, if you or others are very motivated to provide a free conferencing solution for large businesses, go ahead and build it. We always wanted to provide people with a private, self-hosted solution, that’s what motivates us, not helping companies save money. Free Software is about freedom, not ‘free beer’. Freedom is something for people, not companies.

I personally would 10x rather spend time making Talk easier to install and adding features to it that help private users than stop doing that and building a back-end for big companies instead. I do not understand why you care so much about that…

And with regards to strategy, what do you think our strategy is? Making software that costs no money for companies? Not exactly. :stop_sign: Our strategy is to make great software for private users :heart_eyes: and make companies :office: pay for that work, so I have no tears :crying_cat_face: for companies that complain it costs money :money_with_wings: How else could we make Nextcloud better? Eat air and drink angry user posts on our forums? :angry:


That is exactly what Kurento and OpenVidu would be, in line with “not reinventing the wheel”.

The problem is that it isn’t quite useful for more than 5-7 people at best and even then it is problematic. I do not want users (be it home users or small companies that can’t pay a huge sum) to go away to Skype, Hangouts, Meet, Zoom and so on, but it is very hard to archive that with Talk at the moment.

I don’t care much about big corporations (they can bleed money if you want to put it that way), but you still have no offering for small and medium businesses and yes, they need something reliable.

No, but as I said before, not everybody runs a big corporation, and SMB’s are left out in the dark while making up most of the companies out there. An affordable option for them (not free as in “free beer”) would make it easier to get Nextcloud out there to these companies. Big corporations can “bleed”, as I said before, I don’t really care about them.


I agree with you. Another option could be mediasoup which is the backend of multiparty-meeting. It seems very powerful about 130 users per room.