aboutsummaryrefslogtreecommitdiff
path: root/sword2
diff options
context:
space:
mode:
authorTorbjörn Andersson2004-01-13 10:27:13 +0000
committerTorbjörn Andersson2004-01-13 10:27:13 +0000
commit13481b82c45247656b2c453f9e753db9501e4a58 (patch)
tree5a91cb1973cfce49d6ec431cd4d6dadef81f3e53 /sword2
parent8a66ba33cf44600ccc4e2a3267359d2ac952c586 (diff)
downloadscummvm-rg350-13481b82c45247656b2c453f9e753db9501e4a58.tar.gz
scummvm-rg350-13481b82c45247656b2c453f9e753db9501e4a58.tar.bz2
scummvm-rg350-13481b82c45247656b2c453f9e753db9501e4a58.zip
As an experiment, I've added a delay of 0 milliseconds when drawing
interpolation frames. It shouldn't delay anything, but I'm hoping it will allow the other threads some breathing room, which might help bug #875683. Or not. We'll see. svn-id: r12357
Diffstat (limited to 'sword2')
-rw-r--r--sword2/driver/render.cpp5
1 files changed, 5 insertions, 0 deletions
diff --git a/sword2/driver/render.cpp b/sword2/driver/render.cpp
index 5252453004..048f1dc85e 100644
--- a/sword2/driver/render.cpp
+++ b/sword2/driver/render.cpp
@@ -660,6 +660,11 @@ bool Graphics::endRenderCycle(void) {
if (_scrollX != _scrollXOld || _scrollY != _scrollYOld)
setNeedFullRedraw();
+ // This shouldn't delay anything, but might possibly allow the
+ // backend to give the other threads some breathing space, which could
+ // conceivably help against bug #875683.
+ _vm->_system->delay_msecs(0);
+
return false;
}