If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite
Great, now Fedora users can have their updates, while silly little me running gnome on nvidia proprietary maxwell card has the lousiest performance of all DEs out there. Go gnome!
Great, now Fedora users can have their updates, while silly little me running gnome on nvidia proprietary maxwell card has the lousiest performance of all DEs out there. Go gnome!
Why is the performance so bad?
If it's a gnome-only problem why not use KDE, or something else?
Why is the performance so bad?
If it's a gnome-only problem why not use KDE, or something else?
mutter is very very bad. Open some amount of windows and it will be running at 20-15 fps, and it doesn't seem to be maxwell or blob thing, on Intel graphics machines I experience same thing.
mutter is very very bad. Open some amount of windows and it will be running at 20-15 fps, and it doesn't seem to be maxwell or blob thing, on Intel graphics machines I experience same thing.
This isn't my experience at all. On my yoga 2, with just a low power hsw SOC, it runs at 1080 with seven Windows with no noticeable stuttering (when moving between Windows and overview). I used to have a mutter script (iirc, it was, or maybe still is, part of the mutter tree, and used by the developers to test performance) that would open a bunch of Windows, resize them, etc. The point being that mutter is actually pretty good about keeping things smooth.
Great, now Fedora users can have their updates, while silly little me running gnome on nvidia proprietary maxwell card has the lousiest performance of all DEs out there. Go gnome!
what kind of performance problems?
i have 2 systems with different nvidias here and on all of them clutter vblank (vsync) decrease performance. so i have to disable it to can use gnome.
I have this PC configuration: Asus P8B75-V motherboard, Intel core i5-2320, Asus GeForce GTX 750 OC, 8GB RAM. I bought that graphic card couple days ago. As you can see the computer is prety good configuration, and i have problems with almost every desktop environment. When i drag windows around, it lags begind mouse pointer. It happens in unity, cinnamon, gnome 3. In windows i play some games at 1080p, so i would say hardware is not the matter here. Last i tried gnome 3, and switched to ...
This isn't my experience at all. On my yoga 2, with just a low power hsw SOC, it runs at 1080 with seven Windows with no noticeable stuttering (when moving between Windows and overview). I used to have a mutter script (iirc, it was, or maybe still is, part of the mutter tree, and used by the developers to test performance) that would open a bunch of Windows, resize them, etc. The point being that mutter is actually pretty good about keeping things smooth.
Toggling applications view is done and animated by shell, no perfromance problems there, nobody cares about applications view anyway. I was talking about actual window manager duties like moving windows around, doing that peaks CPU usage at near 100% for mutter and X and is ultra-jerky. In contrast, KWIN keeps 60 fps, and CPU usage when moving windows is about 8% and 2% respectively. I should really just record a proper demo.
Toggling applications view is done and animated by shell, no perfromance problems there, nobody cares about applications view anyway. I was talking about actual window manager duties like moving windows around, doing that peaks CPU usage at near 100% for mutter and X and is ultra-jerky. In contrast, KWIN keeps 60 fps, and CPU usage when moving windows is about 8% and 2% respectively. I should really just record a proper demo.
so as i assumed, it seems like the same problem i/we have. just read my post above.
Comment