The Wayland Situation: Facts About X vs. Wayland

[89c51]

Question for everyone else: Would you guys be interested in a systemd / SysV / Upstart comparison as well? I was toying around with the idea in my head, hadn't made up my mind yet.

Nice WL article & yes please do.

[Ericg]

If you want to do that, I can put you in touch with Lennart and Kay to ensure the systemd details are technically correct, if you like. PM or email me.

Thanks Adam, I was planning on contacting one of them anyway but being passed along by you would be nice. Maybe make them more open to talking to me haha. Like I said, it'll be as time permits if I do decide to do it.

[bwat47]

so i do need a freaking EGL to work. So it means endless and countless driver event waking. On my laptop, having kde or xfce with compositing on and no effects makes about 1h15m of battery life. Can you tell me if wayland would provide some mitigation over this issue?

Either your video hardware and/or your video driver is very inefficient if simply enabling compositing significantly effects your battery life. With modern hardware compositing should have no real effect on battery life (and on some hardware using compositing actually IMPROVES battery life, this is how modern intel hardware is designed, the display engine is increasingly decoupled from the rendering so using a pageflipping opengl compositor allows the intel chip to go into deeper power saving states and stay there longer, while also having no video tearing)

If compositing effects your battery life *that* much your hardware/driver is probably more to blame than the compositing itself.

[AdamW]

It is quite obvious it is only for a single axis, not for the total number of pixels. To see this, consider that the old counter was only 15bits long, so it could count 32768 pixels at most. That wouldn't even be enough to hold 800x600(=480000) pixels if it was for the total number, but obviously X11 can support much larger resolutions. Consequently, these counters must be for holding buffer widths and heights, not for the total number of pixels. Think before you answer next time.

The counters referred to ultimately define the limit of how big things can be, but the one is not a direct replacement for the other, as Wayland and X work entirely differently. It does not follow necessarily that the Wayland one is for a single axis just because the X one is.

[DestroyFX]

I want Wayland so badly but

- I play EVE (Not a problem with OSS Radeon but I need to be ablo to load 7-8 clients)
- I mine crypto's (Have 6 7970 in my comp, all miming with love! but I need to be able to reclock GPU Core and MEM very often plus need to monitor the temperature, especially when I will be in Mexico).

So, i'm stuck with X/fglrx

[bwat47]

Question for everyone else: Would you guys be interested in a systemd / SysV / Upstart comparison as well? I was toying around with the idea in my head, hadn't made up my mind yet.

I'd be interested in that

[Ericg]

It is quite obvious it is only for a single axis, not for the total number of pixels. To see this, consider that the old counter was only 15bits long, so it could count 32768 pixels at most. That wouldn't even be enough to hold 800x600(=480000) pixels if it was for the total number, but obviously X11 can support much larger resolutions. Consequently, these counters must be for holding buffer widths and heights, not for the total number of pixels. Think before you answer next time.

Whoops, thanks for the correction Ultima. Definitely wasn't thinking there haha. Sorry about that, just had surgery yesterday so I'm still a little out of it from the pain meds.

[Ericg]

Great articel!

I'd also like to see an init system comparison. Would be cool if you could throw OpenRC into the mix if you feel competent enough to comment on it.
A group of people started maintaining it for Arch Linux as systemd replacement.

This question might be hard to answer from your perspective but do you think that with Wayland/Mir it will/would be easier for AMD/NVIDIA to maintain their Linux drivers? AMD seems to have trouble keeping up with the latest X.Org version from what I know.

If I wasn't able to find enough solid information, I wouldn't do the writeup on the init systems. I wouldn't purposefully put out wrong information.

As far as the AMD/Nvidia issue....I have NO idea why AMD has such problems keeping up with the Xorg ABI (API shouldn't break. ABI does however.) nVidia puts out compatible drivers sometimes before the X server hits final-- they will have a working driver during the development cycle. So Why AMD lags so much I can ONLY assume is incompetence or laziness on their part.

Wayland, for the record, maintains API compatibility from 1.0 on, and Weston maintains API and ABI for all of the 1.0.x cycle (and then 1.1.x, then 1.2.x and so on and so on)

^That was taken from: http://lists.freedesktop.org/archive...er/005967.html

[Ericg]

I can't believe this thread hit 7.5pages in 5.5hrs o_O Thats gotta be some kind of record for these forums haha

[sobkas]

Phoronix: The Wayland Situation: Facts About X vs. Wayland

With the continued speculation and FUD about the future of Wayland at a time when Canonical is investing heavily into their own Mir Display Server alternative, Eric Griffith and Daniel Stone have written an article for Phoronix where they lay out all the facts. The "Wayland Situation" is explained with first going over the failings of X, the fixings of Wayland, common misconceptions about X and Wayland, and then a few other advantages to Wayland. For anyone interested in X/Wayland or the Linux desktop at a technical level, it's an article certainly worth reading!

http://www.phoronix.com/vr.php?view=18763

I would like to mention this bug:

https://bugs.freedesktop.org/show_bug.cgi?id=11227

It can't be fixed within X11.