<freemangordon>
uvos: Maybe it renders fonts, dunno
<freemangordon>
but it seems it is not the reason for the massive fps drop when hildon runs
<freemangordon>
IIUC it is that hildon must copy every offscreen buffer frame, but maybe it does not track the damage correctly
<freemangordon>
and copies the full offscreen buffer every time
<freemangordon>
hmm, even if I stop h-d, scrolling in abook (for example) is still slow
<freemangordon>
could it be that gtk is doing bad things?
rafael2k has quit [Ping timeout: 260 seconds]
rafael2k has joined #maemo-leste
dev has left #maemo-leste [Disconnected: Replaced by new connection]
dev has joined #maemo-leste
dev has left #maemo-leste [#maemo-leste]
dev has joined #maemo-leste
xmn has quit [Ping timeout: 268 seconds]
pere has quit [Ping timeout: 264 seconds]
uvos has joined #maemo-leste
<uvos>
freemangordon: i think gtk2 is just slow at scrolling beacuse it renders eatch frame
<uvos>
qt is 1 hw accelerated and 2 in sw just blits the viewport up a bit and renders the bottom sliver
<uvos>
i doubt this is related to comp.
<uvos>
i mean comp. is slow on x also because of the extra ipc, idk how mutch this adds, but in x the application usualy renders bits to a buffer and hands that to x to render as a pixmap onto its window, where x then maybe also dose some of its own rendering and the buffer is handed to the compositor to compose onto the display
<uvos>
that this is inefficant and makes x a sorta useless step is why wayland got started
<uvos>
maybe instead of trying to make this faster i would be more worth wile to just run in uncomposited mode most of the time
<uvos>
except while displaying tasknav or a notification banner
<uvos>
hildon dosent need to be composited at all
<uvos>
and it almost works
<uvos>
if you force disable comp. on a non-fullscreen window
<uvos>
hildon will switch back as soon as one of those things are needed
rafael2k has quit [Ping timeout: 248 seconds]
<uvos>
the only problem is that the window title bar that hildon renders is missing, im not sure why that, probubbly because its not a real window but is just clutter geometry thats added during compositing, is but there is no reason that could not be a regular x window above the application window.
<uvos>
if we change that (not sure how mutch work in hildon) we could just allways be whithout compositing while there is a application window on top
<uvos>
that saves a huge amount of ram too
<Wizzup>
switching would be much slower though
<uvos>
its ok
<uvos>
really
<uvos>
(try it)
<uvos>
but yes switching causes some extra delay
dev has left #maemo-leste [Disconnected: Replaced by new connection]
dev has joined #maemo-leste
dev has left #maemo-leste [Disconnected: Replaced by new connection]
dev has joined #maemo-leste
pere has joined #maemo-leste
akossh has joined #maemo-leste
Livio has joined #maemo-leste
Livio has quit [Changing host]
Livio has joined #maemo-leste
Livio has quit [Ping timeout: 264 seconds]
akossh has quit [Ping timeout: 246 seconds]
uvos has quit [Remote host closed the connection]
dev has left #maemo-leste [Disconnected: closed]
dev has joined #maemo-leste
dev has quit [Quit: Gateway shutdown]
norayr has quit [Quit: Gateway shutdown]
norayr has joined #maemo-leste
norayr has left #maemo-leste [Error from remote client]
elastic_dog has quit [Ping timeout: 264 seconds]
elastic_dog has joined #maemo-leste
elastic_dog has quit [Quit: elastic_dog]
elastic_dog has joined #maemo-leste
norayr has joined #maemo-leste
akossh has joined #maemo-leste
akossh has quit [Ping timeout: 246 seconds]
xmn has joined #maemo-leste
norayr has left #maemo-leste [Error from remote client]
norayr has joined #maemo-leste
Livio has joined #maemo-leste
Livio has quit [Changing host]
Livio has joined #maemo-leste
xmn has quit [Read error: Connection reset by peer]
xmn has joined #maemo-leste
xmn has quit [Ping timeout: 268 seconds]
LjL-LabRat has quit [Read error: Connection reset by peer]