There's Finally An Experimental Driver For Native Wayland Support Within Wine

[Volta]

Using the same logic pretty much every company falls in that bucket (including AMD/Intel and Microsoft who also contribute to the kernel).

Also such an attitude is frankly immature and stupid since it doesn't get you anywhere.

Everyone can have a reason to hate something. Mine is simple: no open source drivers. It's not stupid, because it prevents me from buying their hardware and suffer from lack of good Wayland support. Furthermore, it seems their prices are way too high in comparison to AMD.

[Volta]

Who wrote the opposite? I wrote that wayland is stable, but that there is still a lot of software that is not compatible with wayland yet and wine is an example. So the experience on wayland at the moment is still incomplete, what is it not you don't understand?

It's complete with XWayland which is part of Wayland, so I don't get your point.

[Weasel]

Not really. With Nvidia being a jackass on not giving XWayland drivers this has to happen. This will allow valve gamescope to be used on Nvidia platforms under X11 more. Remember it needs Wayland support at least to work by some means.

Of course as wine works more straight on Wayland the one of the arguments against wayland will disappear.

Lot of ways you can blame this need to-do this now straight on Nvidia.

It won't. Because Wayland will never support Wine properly, since it's rotten and flawed by design. And I mean actual wine, not crap like proton full of hacks just for games.

[Weasel]

Working on X is waste of time.

But at least it works on X. Wine will never work on Wayland because it's fucked up by design.

And "work" here means, in theory, able to pass all tests, with enough effort implementing it. Not your shitty gaming session.

There's no future when you know you'll never be able to properly implement the Windows API on top of Wayland, literally waste of time, a dead end.

[Charlie68]

It's complete with XWayland which is part of Wayland, so I don't get your point.

The point is that when you use XWayland there is a worsening of performance and memory usage, this as I wrote, may not be a problem on a modern PC, but on older PCs it is, not to mention other side effects that in there are some cases.

[Weasel]

And why is that, exactly?

See my reply 2 posts above.

[NateHubbard]

But at least it works on X. Wine will never work on Wayland because it's fucked up by design.

And "work" here means, in theory, able to pass all tests, with enough effort implementing it. Not your shitty gaming session.

There's no future when you know you'll never be able to properly implement the Windows API on top of Wayland, literally waste of time, a dead end.

I have a feeling this is one of those statements that will look ridiculous in a couple of years when it's proven completely wrong.
I will say though, that according to the patch (which I built and seemed to work fine) there are some situations where things won't work as expected without a wayland extension. The constant need for wayland extensions to do basic things are without a doubt the reason wayland is taking a decade to see any real use.

[duby229]

I have a feeling this is one of those statements that will look ridiculous in a couple of years when it's proven completely wrong.
I will say though, that according to the patch (which I built and seemed to work fine) there are some situations where things won't work as expected without a wayland extension. The constant need for wayland extensions to do basic things are without a doubt the reason wayland is taking a decade to see any real use.

Man, I feel like quoting a movie again... But I won't this time...

The constant need for Wayland extensions is because its scope is too narrow, it'll never be capable of supporting all the features a modern desktop compositor needs. It wasn't designed for modern desktops...

[NateHubbard]

The constant need for Wayland extensions is because its scope is too narrow, it'll never be capable of supporting all the features a modern desktop compositor needs. It wasn't designed for modern desktops...

So what do you think is going to happen? Wayland itself gets replaced?

[oiaohm]

But at least it works on X. Wine will never work on Wayland because it's fucked up by design.

And "work" here means, in theory, able to pass all tests, with enough effort implementing it. Not your shitty gaming session.

There's no future when you know you'll never be able to properly implement the Windows API on top of Wayland, literally waste of time, a dead end.

This is on the incorrect presume that you could implement Windows API on top of X11 correctly. Wine has had virtual desktop and render windows boarders itself options because the reality is you cannot do all with X11 either.

Hidpi monitors people wanting to run old programs want to be able to scale applications up so lie to application that is has 800x600 screen an scale it up wayland version of wine will be able to use gamescope from valve to-do this.

https://www.winehq.org/pipermail/wine-devel/2020-December/178575.html

Also Weasel it would pay to go read this and find out that other than a few unique oddities even with Wayland limitations on screen placement stuff 99% works. This person was not just testing gaming applications either.

Please note the person has basically implemented the exact same method I told you could be done to get around wayland lack of absolute placement that you were like bull crap that cannot work. Sorry it works with some minor limitations that someone may work out a proper solve to. It is very likely that Wine wayland backend with a little more time will be able to pass the complete wine testsuite.

Wayland makes some things harder to implement and requires different thinking but is mostly not impossible. There are a few impossibles with X11 server like not being able to use different color depths on the same screen there are some programs under Windows that do need this that currently are not supported by wine but could be supported by Wayland backend of wine.

The reality the there is stuff of the Windows API that could not be implemented correctly under X11 that will be able to be implemented under Wayland correctly. Scaling support will be better with Wayland backend for Nvidia users due to Nvidia not releasing Xwayland support. Its a true mixed bag problem.

The big thing here Weasel if wine developers do work out a fully functional solution the fucked up by design arguement has got go away replaced purely with its different by design so requiring different design logic.

Heck the fucked up by design at this point has to be taken with a serous grain of salt with what already works with wine on Wayland with quite simple work around to convert from absolute positions of X11 and Windows to relative positions of Wayland. Wayland design differences are showing up to be able to be worked around with very little overhead. This is different to the many things Wayland allows that X11 backend of wine cannot do because X11 protocol does not allow those things at all as this are truly unworkable around limitations or if you do work around them have massive overhead so pointless.

Weasel you were not thinking that X11 is fucked up by design for implementing Windows API as well. The reality between wayland and x11 as wine backends its more of a question where do you want to take your hits with the differences and are these hits at tolerable levels.

I have a feeling this is one of those statements that will look ridiculous in a couple of years when it's proven completely wrong.

Problem is they are already wrong and would have known so if they had bothered reading the link or watched the complete video. As the person doing this development is testing your normal applications as well as games.