Hey, is anyone else experiencing this issue?
When using the background image feature in Nextcloud Talk, there’s a flickering, blurry border around the person. We previously used Google Meet and are now looking for a self-hosted alternative but with this kind of behavior, we can’t use it with our clients.
I only tested it directly on the Nextcloud website and haven’t installed it on our own server yet.
Does anyone have a tip on how to remove the flickering?
Hey @wemakemarketing welcome to the Nextcloud community 
you posted a request in the support category. Unfortunately you ignored the template and a lot of information to help you is missing. Please edit your original post and add all required details like Nextcloud version, webserver type and version, os version, related log file content. Use the support template.
Without additional information the community members cannot help you.
meanwhile check out the docs:
Hey,
thanks for the reply. I ignored the template because I used the nextcloud trial on this page Direkt starten: Ihr kostenloser Sofort-Test - Nextcloud
I thought it is not necessary to type in all those infos because I don’t have them. I am using a macbook pro with the latest os installed if that helps.
Do you need further informations?
I briefly tested it, and Google Meet shows the same effect, although perhaps a bit less noticeable than in Nextcloud Talk.
I’m not sure about the technical reasons behind it, or whether there’s anything that could be optimized. However, it might be worth checking on GitHub to see if an issue has already been reported, and if not, consider opening one.
Hey bb77 thanks for you reply. All I found is this on github that could be similar to my experience: https://github.com/nextcloud/spreed/issues/16008
As I am not really familiar how the github site works, am I right that they are working on the issue and trying to finish the implementation with the next major update for February 28 2026?
Yes, from what I can see (but I don’t have any direct insight into Nextcloud’s development process either), it seems that they have set a milestone for version 33, suggesting that improvements are planned for that release.
Also, from what I’ve gathered, the current implementation appears to be heavily CPU-bound, but more efficient, less CPU-intensive algorithms are available. Google apparently uses one of these better-optimised implementations. Therefore, depending on the hardware, the tearing effect may be more or less noticeable. This also means that the extent to which the effect is noticeable compared to Google’s implementation will likely vary considerably between different devices.
Ok I see, thanks for your answer.
How can I mark this thread as closed?
@wemakemarketing done 