[vdr] VDR client/server (was Re: Developer versions)

2011-01-13 Thread Johan Andersson
Tony Houghton skrev 2011-01-13 18:44: The client-server model is almost essential to me. I wouldn't be interested in a solution that ties me to watching on one PC only and won't let me turn off playback without also preventing background recording. I'd be using mythtv by now if the DVB-S support

Re: [vdr] Developer versions

2011-01-13 Thread Simon Baxter
AIUI VDR generates the OSD as a bitmap no matter which output plugin is used and the player only has the choice of how to overlay it on the video. So getting it rendered by VDPAU would be easy enough, but to upgrade it to HD would probably need some rewriting/patching in core VDR, not just a plugi

Re: [vdr] Developer versions

2011-01-13 Thread Klaus Schmidinger
On 13.01.2011 21:42, Tony Houghton wrote: > ... > AIUI VDR generates the OSD as a bitmap no matter which output plugin is > used and the player only has the choice of how to overlay it on the > video. So getting it rendered by VDPAU would be easy enough, but to > upgrade it to HD would probably nee

Re: [vdr] Developer versions

2011-01-13 Thread Tony Houghton
On 13/01/11 18:57, VDR User wrote: I agree, the less layer upon layer upon layer, the better. I also want to point out that there are a large number of users who have dedicated VDR boxes connected directly to tv's in an htpc environment, using only a remote control to navigate the menus and ssh

Re: [vdr] Developer versions

2011-01-13 Thread VDR User
I agree, the less layer upon layer upon layer, the better. I also want to point out that there are a large number of users who have dedicated VDR boxes connected directly to tv's in an htpc environment, using only a remote control to navigate the menus and ssh for box maintenance. Not computer mo

Re: [vdr] Developer versions

2011-01-13 Thread Tony Houghton
On 13/01/11 16:50, Timothy D. Lenz wrote: And you are back to layer on layer on layer. If someone is going to write something, let it be the plugin that talks to vdr and ffmpeg. Forget xineliboutput, libxine, etc. The more middle ware we can dump the better. So maybe your needs would be best se

Re: [vdr] Developer versions

2011-01-13 Thread Tony Houghton
On 13/01/11 00:11, VDR User wrote: So instead of maintaining just a plugin (which depends on ffmpeg decoding rather then xinelibs decoding), you think maintaining a new player altogether in addition to a plugin that streams data into it? Not to mention forcing VDR into being a backend only. I kn

Re: [vdr] Developer versions

2011-01-13 Thread Gerald Dachs
> oh, so now we need opengl and a compositing manager on top of > everything else? You miss the point. If you would quote mails right you would have noticed that I answered a sentence that told that it doesn't work at all. How did I miss this point? > > have been far from successful. It appears t

Re: [vdr] Developer versions

2011-01-13 Thread Oliver Schinagl
On 01/13/11 15:46, Gerald Dachs wrote: >> >> On 01/13/11 13:31, Rolf Ahrenberg wrote: >>> On Wed, 12 Jan 2011, VDR User wrote: >>> And you get VDR's full osd doing this? >>> FYI, xineliboutput provides three different OSD implementations: > xinelib, composite HUD, and opengl HUD. For example

Re: [vdr] Developer versions

2011-01-13 Thread Timothy D. Lenz
oh, so now we need opengl and a compositing manager on top of everything else? You miss the point. I don't have a problem with vdr being usable as client/server. There are advantages. But it doesn't need to be so complex and it is seperate form the local renderer. On 1/13/2011 7:46 AM, Geral

Re: [vdr] Developer versions

2011-01-13 Thread Timothy D. Lenz
And you are back to layer on layer on layer. If someone is going to write something, let it be the plugin that talks to vdr and ffmpeg. Forget xineliboutput, libxine, etc. The more middle ware we can dump the better. On 1/12/2011 12:15 PM, Tony Houghton wrote: On 12/01/11 18:42, VDR User wrot

Re: [vdr] Developer versions

2011-01-13 Thread Lucian Muresan
On 13.01.2011 13:31, Rolf Ahrenberg wrote: > On Wed, 12 Jan 2011, VDR User wrote: > >> And you get VDR's full osd doing this? > > FYI, xineliboutput provides three different OSD implementations: > xinelib, composite HUD, and opengl HUD. For example the composite HUD > OSD is drawn directly onto t

Re: [vdr] Developer versions

2011-01-13 Thread Gerald Dachs
> > > On 01/13/11 13:31, Rolf Ahrenberg wrote: >> On Wed, 12 Jan 2011, VDR User wrote: >> >>> And you get VDR's full osd doing this? >> >> FYI, xineliboutput provides three different OSD implementations: xinelib, composite HUD, and opengl HUD. For example the composite HUD OSD is drawn directly ont

Re: [vdr] Developer versions

2011-01-13 Thread Michal
On 01/13/2011 03:24 PM, Oliver Schinagl wrote: I'm curious as to how you got the composite and opengl HUD's working, I have been far from successful. It appears to just not work at all :S I've tried composite HUD and it works and looks really nice. The bad is that enabling composite extension

Re: [vdr] Developer versions

2011-01-13 Thread Oliver Schinagl
On 01/13/11 13:31, Rolf Ahrenberg wrote: > On Wed, 12 Jan 2011, VDR User wrote: > >> And you get VDR's full osd doing this? > > FYI, xineliboutput provides three different OSD implementations: > xinelib, composite HUD, and opengl HUD. For example the composite HUD > OSD is drawn directly onto tra

Re: [vdr] Developer versions

2011-01-13 Thread Rolf Ahrenberg
On Wed, 12 Jan 2011, VDR User wrote: And you get VDR's full osd doing this? FYI, xineliboutput provides three different OSD implementations: xinelib, composite HUD, and opengl HUD. For example the composite HUD OSD is drawn directly onto transparent window located exactly over the (xine-lib