Mark Shuttleworth Declares Mir A Performance Win

[przemoli]

Its FLOSS.

Fork is our sun. Deal with it.

(Fork as any code sharing that is done according to license)

And working together on something better is our air. Deal with it.



(First relate to notion that Mir copying code from Wayland is wrong... Second relate to idea that Mir reusing Mesa work could be wrong... Hilarious.)

[ChrisXY]

Mir is currently in Ubuntu 13.10, working in action. You can download the daily build and use it NOW. Where is Wayland? I hear talk for years but nothing in action.

Maybe you're not looking hard enough?

http://www.youtube.com/watch?v=9_uRUIctGd0

http://www.youtube.com/watch?v=-V8i8zZPzbU

[e8hffff]

It's a great opportunity for the Wayland developers to create their own distribution of Linux and go head to head with Canonical. A means to show what they've got, balls or talk.

I'll be staying with Kubuntu and maybe move to Ubuntu with KDE installed using XMir if the Kubuntu people are so ignorant.

[dh04000]

It's interesting to see that the wayland guys have been very careful about doing things properly and not pushing beta software out to users, as has happened in some transitions on the linux stack before (like pulseaudio), but then canonical shows up and does the good old "here goes something" bit.

Honestly I'm torn. I use Ubuntu, and plan on continuing doing so, but I think Mir is just the result of canonical devs just giving into NIH, while Wayland seems to be the path forward. But then, hurd was proper and linux was something. Who knows what will happen.

If you want to stay in the ubuntu ecosystem, wayland will be availabe to install when it is ready for consumption(hits debain testing). So you don't have to make a hard decision, install both when wayland become available and see which one you like, all staying on the Ubuntu ecosystem.

Its exciting to see how things will turn out. I like your Hurd v Linux anology. Hopefully though, both will survive, instead of the one being successful and the other nonexistent, like your example.

[jrch2k8]

The wayland guys don't have the power to do anything. No other distro cares enough about Wayland to beta test it for them, so of course it hasn't been pushed out. Canonical on the other hand, has a time frame they must meet, and they care deeply about Mir, so they're going to do whatever it takes. They think it will be past the beta stage by the time Ubuntu 13.10 is released, but they still need to have it tested in real world conditions before the LTS in the spring.

I appreciate what the wayland community is doing, but don't give them credit for making a decision they never had the power (and by extension, the temptation) to make.

ok lets get the hard to miss facts that if you actually read the code would OVBIOUS before answer your post

1.) Xmir don't test at all anything that a native mir apps will need, the only thing Xmir do is provide surfaces to push them back/forth to Xserver and very basic operations like input, scaling and location of surfaces. this wayland do since ages ago
2.) Mir technically speaking is very far away of wayland capabilities[like color accurate render, frame sync, CSD, SSD, multiformat surface, multi-input, subsurfaces, software only allocations, agebuffer EXT, overlay, color type convertion, video surfaces,etc], so basically Mir is in the ages of wayland 0.20 or something around that age, i mean basically Xmir exploit all the working feature available in Mir today
3.) Xmir is not working correctly in some desktops like KDE SC and i saw in a mailing list EFL is not a happy camper with Xmir either

if you need evidence find someone who can understand C and C++ and ask him to read both codebases

ok, now to your post

1.) Xwayland is not even considered stable[canonical wonderboys don't care it seems], for wayland this is nothing beyond a fast hacky experiment to test if it was possible[very clearly stated, check mailing list] but again canonical ripped of what it could to show some PR, Proper Xwayland will come in xorg-server 1.15+ they need kernel side fixes to do it properly
2.) Wayland stated again it should never be used until 1.0 was reached[again mailing list access is free you know]
3.) Wayland and everybody else[Mplayer/Gtk/Qt5/EFL/SDL/KDE/GNOME/e17/etc] except canonical wonderboys are focused full steam ahead in native ports instead of make PR with some technically useless demo or push alpha software to users
4.) Qt5, GTK , SDL, EFL apps already runs very good in wayland natively and Gnome/e17 should work natively in wayland this year tops and KDE prolly will in Q1 2014, like i said before wayland ppl don't make PR they port
5.) Wayland support out-of-the-box will happen in Q1 2015 but will be available as a choice backend for anyone who wants testing it way before that, so they can iron out any issue before replace Xorg[unlike canonical this guys are proffesionals and won't push it to replace Xorg until it has enough testing], this was accorded between many parties and again mailinglist access is free
6.) Wayland don't need PR or an specific distro, Wayland is the linux default choice backed by freedesktop.org and Xorg foundation that will use any GNU/Linux distro on earth[keeping Xorg as fallback] from now on except for canonical OS until they can't take the punishment anymore[keep ports of every major GUI software downstream for Mir compat]
7.) at last if i want i can make KDE SC 4.11 on Xwayland in arch or gentoo but why use alpha software if i can wait some months to have full native DE to test it properly? and as a fact i will only needed it for GTK+-2.x applications if at all

conclusion literate before post if you don't wanna be seen as a religious fanboy

[TheOne]

It seems that Canonical developers didnt show the phoronix benchmarks to Mark and he is having a placebo effect.

[Siekacz]

ok lets get the hard to miss facts that if you actually read the code would OVBIOUS before answer your post

1.) Xmir don't test at all anything that a native mir apps will need, the only thing Xmir do is provide surfaces to push them back/forth to Xserver and very basic operations like input, scaling and location of surfaces. this wayland do since ages ago
2.) Mir technically speaking is very far away of wayland capabilities[like color accurate render, frame sync, CSD, SSD, multiformat surface, multi-input, subsurfaces, software only allocations, agebuffer EXT, overlay, color type convertion, video surfaces,etc], so basically Mir is in the ages of wayland 0.20 or something around that age, i mean basically Xmir exploit all the working feature available in Mir today
3.) Xmir is not working correctly in some desktops like KDE SC and i saw in a mailing list EFL is not a happy camper with Xmir either

if you need evidence find someone who can understand C and C++ and ask him to read both codebases

ok, now to your post

1.) Xwayland is not even considered stable[canonical wonderboys don't care it seems], for wayland this is nothing beyond a fast hacky experiment to test if it was possible[very clearly stated, check mailing list] but again canonical ripped of what it could to show some PR, Proper Xwayland will come in xorg-server 1.15+ they need kernel side fixes to do it properly
2.) Wayland stated again it should never be used until 1.0 was reached[again mailing list access is free you know]
3.) Wayland and everybody else[Mplayer/Gtk/Qt5/EFL/SDL/KDE/GNOME/e17/etc] except canonical wonderboys are focused full steam ahead in native ports instead of make PR with some technically useless demo or push alpha software to users
4.) Qt5, GTK , SDL, EFL apps already runs very good in wayland natively and Gnome/e17 should work natively in wayland this year tops and KDE prolly will in Q1 2014, like i said before wayland ppl don't make PR they port
5.) Wayland support out-of-the-box will happen in Q1 2015 but will be available as a choice backend for anyone who wants testing it way before that, so they can iron out any issue before replace Xorg[unlike canonical this guys are proffesionals and won't push it to replace Xorg until it has enough testing], this was accorded between many parties and again mailinglist access is free
6.) Wayland don't need PR or an specific distro, Wayland is the linux default choice backed by freedesktop.org and Xorg foundation that will use any GNU/Linux distro on earth[keeping Xorg as fallback] from now on except for canonical OS until they can't take the punishment anymore[keep ports of every major GUI software downstream for Mir compat]
7.) at last if i want i can make KDE SC 4.11 on Xwayland in arch or gentoo but why use alpha software if i can wait some months to have full native DE to test it properly? and as a fact i will only needed it for GTK+-2.x applications if at all

conclusion literate before post if you don't wanna be seen as a religious fanboy

And yet Wayland cannot display simple OpenGL apps, wheter running in XWayland or not. Mir can.

[riklaunim]

Canonical has a deadline so some will also want a race between it and Wayland. (like "how cool we are, we released before Wayland" and so on).

Also do we have computers that can't handle a desktop? Display server performance is rather illusionary term - it either works, or not. You don't see a difference between 200 FPS or 150 FPS for some drawing operation etc. when in normal conditions max of 50 FPS is achieved as there is nothing else to draw and so on.

[e8hffff]

Also do we have computers that can't handle a desktop? Display server performance is rather illusionary term - it either works, or not. You don't see a difference between 200 FPS or 150 FPS for some drawing operation etc. when in normal conditions max of 50 FPS is achieved as there is nothing else to draw and so on.

Multitasking on larger screens is why people aim for higher frame rates in gfx cards.

[rockmen1]

So Mr. Mark Shuttleworth your team forked wayland's code base and sync to it one a while in the future, then make some modification to suit your needs, and this is called "evolve faster than the competition"?