Experimental Wayland Support For Wine Now Sees More Functionality Working

[RahulSundaram]

That's both splitting hairs and not what anyone is claiming. What the Devs claimed is that the X11 architecture was garbage in quite the laundry list of ways and that one of the advantages (not goals... advantages) of Wayland was that because it fixed those things it would be faster

What people are saying is that the stated advantage has not materialized.

Alright, let's go with your restated claim here. Which Wayland developer said performance would be an advantage?

[Luke_Wolf]

Alright, let's go with your restated claim here. Which Wayland developer said performance would be an advantage?

You do realize you're asking me to remember specific names from threads that descended into complete flamewars from around a decade ago, right? And I just checked and the forum archive only goes back to 2018 when Michael rebuilt the site, so it's not as if I could "prove" it to you anyway. However for some reason Daniel Stone is coming to mind.

[RahulSundaram]

You do realize you're asking me to remember specific names from threads that descended into complete flamewars from around a decade ago, right? And I just checked and the forum archive only goes back to 2018 when Michael rebuilt the site, so it's not as if I could "prove" it to you anyway. However for some reason Daniel Stone is coming to mind.

I am just looking for any kind of reference anywhere, not limited to this site at all.

A name is something. I will reach out to him.

[oiaohm]

I just would like to point out that this argument that Wayland support somehow will lead to improved performance simply isn't true. Improved performance hasn't been demonstrated at all, not even once in all these years. It's just not true.

https://www.phoronix.com/scan.php?page=news_item&px=GNOME-Xorg-Wayland-AMD-Renoir

The claim that it has not been demonstrated at all is incorrect. Tesseract benchmarks is one of those cases. There are other targeted benchmarks that have also shown wayland being better.

So different benchmarks demonstrate that wayland performance could be better than X11 server solution. Key word is could. There are a lot of moving parts. The small performance advantage wayland zero buffer copy stuff gives can very quickly be consumed by other areas being poorly optimised.

[mbrf]

Again, I don't remember specifics, but during the development of Wayland, I know that I've seen a talk by one of the maintainers of X11, talking about how inefficiently it does things, and how it's a mess that no-one has any overview over anymore.
During this time, X11 got a lot of work done, to play nice with Wayland.
I vaguely recall statements, during this time, that:
X11 Vs Wayland, in single window situations, that Wayland should have much less overhead .
That as amount of windows increased, that each Wayland window would require close to the same amount of resources as the first window, while X11 remained somewhat stable in the required resources, as window count increased .
And finally, that X11 quickly was less resource hungry, than 2 or more XWayland windows.

I'm also nearly certain that there later were benchmarks showing the cycles it took drawing the different scenarios, the time it took, and time that was spent just waiting. But I haven't got a clue what to search for, to find it, but I'll give it a go later.
It might only have been projections based on the knowledge of X11, Vs the idea behind Wayland, but I really don't think that it was.

And while I don't think that performance was ever the goal, but an expected benefit under a certain amount of windows, cleaner and more organised and better documented code, was. Perfect frames was. Improved security, was.

[smitty3268]

I am just looking for any kind of reference anywhere, not limited to this site at all.

A name is something. I will reach out to him.

A lot of the discussion by devs back then was just noting you could reduce IPC by integrating the compositor, removing X roundtrips, and various other X overhead. It was all more in the context of optimizing toolkit drawing, though, never anything about making a 3d game run faster and nothing ever really even saying it would be visibly noticeable. Just pointing out that there were many things that could be optimized better.

The phoronix forum trolls saw the word "faster" though, and it became something that people would copy/paste into flamewars everywhere without any kind of context associated with it.

So yes, there was some discussion about it from devs, but nothing resembling what the talking points here eventually became.

[aufkrawall]

you are basically saying that x11 is fast, but only if you run fullscreen. well, it means that in general x11 is slow. and why would you be forbidden to play gpu demanding titles in windowed mode?

It is not slow, there is just no equivalent to dmabuf present (Which is still hardly a thing on Wayland these days?) so you have to run through compositor rendering.
Or you suspend compositing and live with tearing (or enable TearFree, though this adds lag if there was tearing without it).
Reminder: You currently can't play games that tax out the GPU without nasty stutter at all in a Gnome 3.38.3 Wayland session. While Xorg has stupid limitations that can't be overcome, Wayland compositors still need to overcome a lot of issues too.

[duby229]

moron, how improved performance in wayland could be demonstrated without wayland support(subj is about adding wayland support in future)?

Moron, its already been -over- 13 years... Just -exactly- how many -more- years should it be???

[duby229]

https://www.phoronix.com/scan.php?page=news_item&px=GNOME-Xorg-Wayland-AMD-Renoir

The claim that it has not been demonstrated at all is incorrect. Tesseract benchmarks is one of those cases. There are other targeted benchmarks that have also shown wayland being better.

So different benchmarks demonstrate that wayland performance could be better than X11 server solution. Key word is could. There are a lot of moving parts. The small performance advantage wayland zero buffer copy stuff gives can very quickly be consumed by other areas being poorly optimised.

Considering that your own damn link proves you wrong....

[oiaohm]

Considering that your own damn link proves you wrong....

Tesseract OpenGL game wayland in fact wins over X11 and its repeatable.

I stated that its particular benchmarks that show it. Sorry the link I gave did not prove me wrong just proved you did not read. I started Tesseract benchmarks.

The possible advantage of Wayland design is very small we are talking a somewhere between 1-5 frames per second. Is this going to be a human noticeable difference no its not. Is it possible for X11 solution to be sightly losing all the time to a Wayland solution if the Wayland solution is well made absolutely yes.

duby229 big thing here is your claim that it never been demoed in benchmarks that claim is false. The advantage of wayland design can be seen with the right benchmarks.

Please note gnome is not a well made wayland compositor in that those benchmarks. Really that link I gave really shows few how design in the compositor with wayland can totally destory the advantage of the wayland design even in that case particular benchmarks will show the advantage is there, Finally the advantage of wayland over X11 if everything is good is not going to be a major deal breaker.