Mir Now Supports XDG Shell Stable

[phoronix]

Phoronix: Mir Now Supports XDG Shell Stable

Canonical developers continue working on advancing the Mir display server's support for Wayland...

http://www.phoronix.com/scan.php?page=news_item&px=Mir-XDG-Shell-Stable

[andre30correia]

it was a shame linux comunity don't put efforts in mir and not in wayland, my bq aquaris runs much faster with ubuntu than with android. A really great project who advanced a great speed for mobile tablet and desktop/laptop with things usersp need not like wayland who is turned in new x11 with the same old guys who think are in 90's or something. Wayland performance even with native apps is really poor after so many years

[dkasak]

Wayland performance even with native apps is really poor after so many years

Stats? It's faster for me.

[profoundWHALE]

Stats? It's faster for me.

Why would you ask someone else to back up their claim with data, only to follow up with anecdotal evidence?

[dkasak]

Why would you ask someone else to back up their claim with data, only to follow up with anecdotal evidence?

To prove the point. Think about it ... ... ...

[jo-erlend]

it was a shame linux comunity don't put efforts in mir and not in wayland, my bq aquaris runs much faster with ubuntu than with android. A really great project who advanced a great speed for mobile tablet and desktop/laptop with things usersp need not like wayland who is turned in new x11 with the same old guys who think are in 90's or something. Wayland performance even with native apps is really poor after so many years

Speed would never be the difference between Mir and Wayland. Both would be expected to be slower than X.org and Wayland would be expected to be faster than Mir. The benefit I see in the old Mir over Wayland, is that Mir is designed to do server-side buffer allocations, which allows the display system to throttle and steal known unnecessary memory from other apps. This would primarily be a benefit to embedded systems like phones, where RAM is a very serious issue for physical reasons but would also allow for the enforcement of serious programming. I mean it _is_ stupid to decode and draw video that will never ever be observed by anything except the CPU and GPU. A video player should simply not be allowed to decode and draw video when it's a known fact that there will not be any output.

I remember central people in the Wayland design crowd, saying Mir could just use a protocol extension to make Mir a buffer-controller while being compatible with Wayland. I don't know if that's happening with Mir. That would be significant news for all of Wayland. If that's what they're doing, then I think that would be very exciting news for the whole community. Nobody has ever tried to defend the idea that all graphics should belong to the clientside. There are good things to say about governed display if we can get past the politics.

[TheBlackCat]

Both would be expected to be slower than X.org

Why?

The benefit I see in the old Mir over Wayland, is that Mir is designed to do server-side buffer allocations, which allows the display system to throttle and steal known unnecessary memory from other apps. This would primarily be a benefit to embedded systems like phones, where RAM is a very serious issue for physical reasons but would also allow for the enforcement of serious programming.

Wayland supports server-side buffer allocations, it just doesn't enforce them.