Ubuntu Unity Proves Very Slow To KDE, GNOME, Xfce, LXDE

[bug77]

Normally, a window writes into an offscreen framebuffer of the size of the window, which is later copied onto the screen at the position of the window. If the compositor can properly detect a fullscreen window, it can skip the copying and render to the screen directly. It's only the copying that makes the difference.

Ok, that somewhat resembles double buffering (except that this time it's per window). But if the WM only has to draw one "window" and then copy it to screen, is that enough to explain the rather sizeable gap in Michael's graphs?

[LinuxAffenMann]

"Ubuntu Unity Proves Very Slow To KDE, GNOME, Xfce, LXDE" - in games and synthetic benchmarks, OH NOES!

#lol@article

[mrugiero]

As much as I don't want to believe it. Unity is by far slower than Windows. It also takes forever to boot. There seems to be something hanging between login and logout that takes about 60 seconds to get passed it. The unity panel itself lags when scrolling up and down the menus and the dash menu has a big delay when opening. It feels like Windows 7 when you got infected with a bunch of malware 0_o

Ubuntu and Unity aren't the same thing.

[ruinairas]

Unity on AMD FX 4100 feels sluggish compared to even Windows.

As much as I don't want to believe it. Unity is by far slower than Windows. It also takes forever to boot. There seems to be something hanging between login and logout that takes about 60 seconds to get passed it. The unity panel itself lags when scrolling up and down the menus and the dash menu has a big delay when opening. It feels like Windows 7 when you got infected with a bunch of malware 0_o

[Teho]

Well, then, that's what the guy (and I) wanted to know. Do you happen to know how is it done?

Nope. I guess you could ask "mgraesslin" who is the lead developer of KWin and made a post in the "KDE's KWin To Receive Performance Improvements" topic.

[johnc]

I love the screencap of the new Ubuntu login session scrolling right off the screen. I'm serious when I say: I don't think they test this stuff.

So can we conclude that Cinnamon would be similar to GNOME Shell in terms of performance?

[mrugiero]

That's not how it works. Only the full screen window and windows beneath it lose the composition and it doesn't seem to affect the other screen at all. I'm doing that all the time because the only way I can get tear free video on my external monior is by using MPlayer without compositing with VDPAU output . I'm not sure how it affects performance though.

Well, then, that's what the guy (and I) wanted to know. Do you happen to know how is it done?

[Teho]

If you use dual head, use a fullscreen window in one screen. Then, if you choose to suspend effects when fullscreen, you have the other head without visual effects as well, giving you a less appealing desktop. If you choose to not suspend, well, you get the Unity-like performance in the fullscreen app. Beats me what should happen if using two fullscreen apps, one in each screen.

That's not how it works. Only the full screen window and windows beneath it lose the composition and it doesn't seem to affect the other screen at all. I'm doing that all the time because the only way I can get tear free video on my external monior is by using MPlayer without compositing with VDPAU output . I'm not sure how it affects performance though.

[kraftman]

Wasn't L4D2 faster on Ubuntu than on Windows?
Didn't that Ubuntu use Unity? [http://blogs.valvesoftware.com/linux.../#comment-4328]

Just imagine a "KDE SC 4.9 - Suspended" ? so much more frames per second ?

Exactly. I'd love to see benchmarks with KDE compared to OS X and Windows.

[mrugiero]

KDE works flawless with Dual-Monitor: my System http://www.sysprofile.de/id123113.

He didn't talk about KDE itself not working with dual head, but the suspend feature. And I believe it's pretty obvious what he means. If you use dual head, use a fullscreen window in one screen. Then, if you choose to suspend effects when fullscreen, you have the other head without visual effects as well, giving you a less appealing desktop. If you choose to not suspend, well, you get the Unity-like performance in the fullscreen app. Beats me what should happen if using two fullscreen apps, one in each screen.

should valve work with unity/compiz-team to help them improving the perfermance of unity/compiz?
that will affect so much their gaming experience on linux/ubuntu

Why should they? Be happy they support the OS, I don't think they help Microsoft improving Aero.

I wonder why the full screen tweaks actually works. Since there is no more desktop to render, hence nothing to decorate, reason says there should be no performance penalty in the first place.

Because of how compositing works. When using a compositor, you use a buffer for every window, and usually update the content of them all (I'm not completely sure of how this is done, though). This is actually needed for most of the effects, since some of them interacts with the window behind (translucency and shadows, for example). This means you work with the windows which are not being displayed too, so no desktop to render becomes false.

Normally, a window writes into an offscreen framebuffer of the size of the window, which is later copied onto the screen at the position of the window. If the compositor can properly detect a fullscreen window, it can skip the copying and render to the screen directly. It's only the copying that makes the difference.

I assume you know more than I know, since I'm just an amateur reader. Doesn't it use a painter's algorithm? Doesn't the fullscreen mode also allow you to avoid that too?