Wayland-Based Chromium Browser Released

[gufide]

Yeah. The big feature for end users is a tear free environment.
There should be a small performance increase as well.

And it reduce greatly input latency because there is no Vsync and no need to an additional opengl compositor on top of the display server because the compositor IS the display server.

[dee.]

Go watch a youtube video and try to scroll the browser page, if you want to see a concrete example of the benefits of Wayland.

Also this is a good reference:

http://www.phoronix.com/scan.php?page=article&item=x_wayland_situation&num=1

[ua=42]

Yeah. The big feature for end users is a tear free environment.
There should be a small performance increase as well.

[AnonymousCoward]

For the end user, that means nothing new per se, but possibly more things or better quality things in general (in the future.

Well, I'd say a tear-free desktop without having to slap this compositor extention hack somewhere into the stack like with X is kindof a big thing for end users.

I somewhere heard that Waylands motto is "every frame is perfect" which is something that sounds pretty cool and desireable for end users.

[erendorn]

I?m curious about the amount of work needed to port an application to use Wayland. A team at Intel needed 2 months to partially port Chromium which already had an abstraction layer?

Also I never remember what problems Wayland is supposed to fix. I went to their web site to read some doc two days ago but nothing sticked in my mind.

Wayland mostly helps on the developer side, and actually mostly the system or graphic stack developer side.
For the end user, that means nothing new per se, but possibly more things or better quality things in general (in the future. right now, you'll have as much things, but buggier, until the technology gets fully mature).

For example, the wayland back-end for the R-Pi uses the GPU's hardware compositor, with an incredible boost in performance. That's something that cannot be done with X.
It is this kind of little things.

[valeriodean]

I?m curious about the amount of work needed to port an application to use Wayland. A team at Intel needed 2 months to partially port Chromium which already had an abstraction layer?

Also I never remember what problems Wayland is supposed to fix. I went to their web site to read some doc two days ago but nothing sticked in my mind.

Maybe the web browser is not properly the most correct example, because it touches one of that particular case where subsurfaces are needed.
Other than that, there is an additional difficulty in case the browser's toolkit has not been ported yet (I don't remember if chromium uses gtk+ 2 or what).
I guess that, excepts for video player (maybe another subsurfaces case), the porting effort from X to Wayland should't be so hard.

[YoungManKlaus]

I?m curious about the amount of work needed to port an application to use Wayland. A team at Intel needed 2 months to partially port Chromium which already had an abstraction layer?

Also I never remember what problems Wayland is supposed to fix. I went to their web site to read some doc two days ago but nothing sticked in my mind.

well, a team can have many responsibilities ... I also have some (rather small) "pet" features in the pipeline here at work which started a month ago but got burried under other work because of other priorities (and eg. get 3-4 hours a week tops). As for advantages, the biggest one imo is simply that all the legacy x11 features which are no longer used by real-world apps are gone which leads to a way smaller codebase (which has thus - statistically speaking - less bugs). Jeah, the architecture is nice and has some nifty features (tear-free rendering as an example), but I dont worry about that so much as about a huge-ass piece of code that has dubious quality (I remember reading about some x.org releases not even compiling) and runs in kernel-mode at the same time.

[stqn]

I?m curious about the amount of work needed to port an application to use Wayland. A team at Intel needed 2 months to partially port Chromium which already had an abstraction layer?

Also I never remember what problems Wayland is supposed to fix. I went to their web site to read some doc two days ago but nothing sticked in my mind.

[YoungManKlaus]

And suddenly...

Wayland is a workable alternative for probably 80% of the typical users day-to-day work

[tarceri]

Firefox can do a lot quickly on issues they care about, but linux integration has never been one of those. Anyone expecting quick Wayland support from Firefox needs to lower their expectations.

Looking at the blocker bugs to the Firefox bug to add wayland support it looks like performance concerns seem to be holding back some of the work to switching to using Cario over Xrender.

See: https://bugzilla.mozilla.org/show_bug.cgi?id=738937

Aside from that looks like there has already been a bit of work done towards replacing GLX with EGL see: https://bugzilla.mozilla.org/show_bug.cgi?id=788319

So basically things are moving slowly but there seems to be no real push to get Firefox working on Wayland.