The team behind the Matrix open standard and real-time communication protocol has announced the release of its second major version, bringing end-to-end encryption to group VoIP, faster loading times, and more.
For me that has always been a UI issue. Many chat platforms will show you your message even if it hasn’t been sent yet, but many Matrix clients will wait for the message to be confirmed at least by your own home server before they’re added to the chat list. Others do add a local message, but have such complex interaction logic that there’s a noticeable delay between hitting send and your message appearing.
I must say clients have gotten better over the years, though Element on desktop still isn’t entirely there in terms of responsiveness for me.
Latency between Matrix servers running Synapse (which are most servers) is quite high, up to a few seconds if you haven’t used a particular room in a while, but it’s not much higher than many other platforms that just hide the latency better. Latency on the same server is almost negligible in my experience, unless the server is severely overloaded.
Latency between servers written in faster languages (such as Dendrite and Conduit) is much lower, although most feature development doesn’t happen in those servers, and they’re constantly playing catchup with Synapse. I understand that the Matrix folks are struggling to make money already, but their choice to run their server in Python of all languages is hurting Matrix usability.
I’ve never encountered that, even when messaging to the other side of the world. There’s something more going on when it takes that long, but I don’t think that’s normal.
For me that has always been a UI issue. Many chat platforms will show you your message even if it hasn’t been sent yet, but many Matrix clients will wait for the message to be confirmed at least by your own home server before they’re added to the chat list. Others do add a local message, but have such complex interaction logic that there’s a noticeable delay between hitting send and your message appearing.
I must say clients have gotten better over the years, though Element on desktop still isn’t entirely there in terms of responsiveness for me.
Latency between Matrix servers running Synapse (which are most servers) is quite high, up to a few seconds if you haven’t used a particular room in a while, but it’s not much higher than many other platforms that just hide the latency better. Latency on the same server is almost negligible in my experience, unless the server is severely overloaded.
Latency between servers written in faster languages (such as Dendrite and Conduit) is much lower, although most feature development doesn’t happen in those servers, and they’re constantly playing catchup with Synapse. I understand that the Matrix folks are struggling to make money already, but their choice to run their server in Python of all languages is hurting Matrix usability.
deleted by creator
I’ve never encountered that, even when messaging to the other side of the world. There’s something more going on when it takes that long, but I don’t think that’s normal.