Web UDP - public demand

[mrmaxm]

Many of you know games like agar.io, slither.io and many others. Most of them are implemented on top of WebSockets, which is based on TCP. Due to that pretty much all of them suffer from some latency issues making them less responsive and limiting over what multiplayer games can be implemented in the web today.

So here is public demand for Web UDP: https://github.com/Maksims/web-udp-public

Which is collaborative effort to shape requirements by developers as well as awareness about need of UDP, so to motivate W3C members and browser vendors for work on specs and implementations that will allow use of UDP in server-client cases.

 

Please contribute, participate and share to raise awareness and get the momentum.

[mattstyles]

I saw a tweet from you this morning about this, sounds great in principle, that small section on security concerns is massive though! That's got to be the main sticking point, not that I know a great deal about it.

[Antriel]

This guy also tries to push something like that forward: https://gafferongames.com/post/why_cant_i_send_udp_packets_from_a_browser/

[caymanbruce]

I agree with you. I have first hand experience and I can see how painful the realistic is. I wholeheartedly support your demand.

[b10b]

Netcode.io is a solution that uses what currently exists to design around the problem - it's great to see such an implementation approaching a mature state.  But the issues remain: use case is marginal, support is spotty.

Will that change?  I am sceptical there exists a significant untapped audience of gamers who seek UDP experiences in their web browser.  Instead I see gamers who are conditioned to install for optimal experience (apps) or accept latency for convenience (web).  If there exists more benefits for a wider group of users then perhaps gathering data and adding that to the proposal would be beneficial?