summaryrefslogtreecommitdiff
path: root/README-VDPAU
diff options
context:
space:
mode:
Diffstat (limited to 'README-VDPAU')
-rw-r--r--README-VDPAU40
1 files changed, 40 insertions, 0 deletions
diff --git a/README-VDPAU b/README-VDPAU
new file mode 100644
index 000000000..c664e36af
--- /dev/null
+++ b/README-VDPAU
@@ -0,0 +1,40 @@
+FAQ:
+
+Q:
+ Why my file plays fine with mplayer-vdpau and not with xine-vdpau?
+A:
+ We are not using the nvidia's libavcodec patch.
+ We are writing decoders from scratch.
+ So don't expect them to be as mature as ffmpeg ones. Not yet.
+
+Q:
+ Why mpeg2 doesn't use less cpu than software decoder?
+A:
+ Because at that moment it does a lot of memcpy. This will be fixed soon, but that's not
+ a priority. Stability is our focus.
+
+Q:
+ Is deinterlacing working?
+A:
+ Yes. It's already quite good (doing 50i->50p), but could even be better in the future.
+
+Q:
+ How do i get it working with VDR, Kaffeine, whatever.
+A:
+ Ask VDR, Kaffeine, whatever developers.
+ (Note: for kaffeine you are lucky, i'm going to tell you the tip.
+ Build kaffeine like that: ./configure --without-xcb && make && make install)
+
+Q:
+ How can i contact you?
+A:
+ IRC: #xine-vdpau on freenode
+ MAIL: http://lists.kafic.ba/mailman/listinfo/xine-vdpau
+ Eventually, nvnews.
+
+Q:
+ What information do I need if there's a problem?
+A:
+ Run 'xine --verbose /path/to/a/working/sample' (or equivalent for your
+ chosen front end). You will probably be asked for this anyway, so it
+ helps to have it handy.