Re: [vdr] vdr 1.7.4 HD-Recording not working

2009-03-12 Thread Artem Makhutov
Hello,

Oliver Bardenheier schrieb:
> I just tried to record some HD but found that VDR is crashing after a few
> secs.
> Watching HD is no problem, but when going on record VDR exits.
> First I though it was noad related, but I switched off any unneded plugin
> and noad, still crashing.

I have no problems in recording Arte HD at all. Everything works fine 
for me.

Regards, Artem

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [VGA sync field] question to fb-radeon-intel.patch

2009-03-12 Thread Paul Menzel
Am Donnerstag, den 12.03.2009, 18:31 +0100 schrieb Thomas Hilber:
> On Wed, Mar 11, 2009 at 06:15:04PM +0100, Paul Menzel wrote:
> > 1. In your patchset the patch fb-radeon-intel.patch [1] does include
> > changes to drivers/video/intelfb/intelfbhw.c although in the install
> > instructions [2] it is stated that intel drivers do not need to be
> > modified.
> 
> sorry, install instructions are not up to date for the Intel version
> of the patch.
> 
> > 
> > The changelog in 0.11 includes these two items.
> > 
> > - patch against intelfb (kernel 2.6.26) to allow for PAL/SCART
> >   video timings. You now can use a regular SCART CRT as display
> >   for linux console.
> > - fixed a bug in intelfb initialization which sporadically
> >   setup video timing with weirdous values
> > 
> > I guess the installation instructions need to be updated for release
> > 0.11.
> 
> right. as said installation instructions have not been touched since
> a while. I just included the plain Intel patches into the package.
> There was not yet too much interest in the patches anyway. So I 
> did not want to waste my time for documentation:-) 
> 
> > If this is correct I would update the instructions.
> 
> thank you for that.

Done (commit 89ddb9af67be949be3be01f795f964926728f45a). Not much, but I
hope alright.


Thanks,

Paul


signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil
___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [VGA sync field] question to fb-radeon-intel.patch

2009-03-12 Thread Thomas Hilber
On Thu, Mar 12, 2009 at 07:36:53PM +0100, Torgeir Veimo wrote:
> Did you look into making your approach work with DirectFB yet?

I don't think it's a big deal to port the Intel version of the
FRC patch to DirectFB. 

But sorry, I don't use DirectFB. I can't see much advantage over a plain
Xserver unless running something like an embedded system.

- Thomas


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [VGA sync field] question to fb-radeon-intel.patch

2009-03-12 Thread Thomas Hilber
On Thu, Mar 12, 2009 at 07:36:01PM +0100, Paul Menzel wrote:
> I hope, you have seen my message to linux-fbdev-devel [1]. I will try to
> do what Krzysztof suggested, but will have to read more about that.
> Maybe we get it into Linux kernel 2.6.30.

yeah, I've seen that. 

> Should we continue discussion about this on linux-fbdev-devel, that
> means, can subscribe there?

I subscribed to linux-fbdev-devel. But haven't had the time to adapt the
patch as requested.

- Thomas


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [VGA sync field] question to fb-radeon-intel.patch

2009-03-12 Thread Torgeir Veimo

On 12 Mar 2009, at 19:08, Thomas Hilber wrote:

> I forgot to say patching the intelfb is only needed if you even want  
> to
> run the linux console in VGA2SCART mode.


Did you look into making your approach work with DirectFB yet?

-- 
Torgeir Veimo
torg...@pobox.com





___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [VGA sync field] question to fb-radeon-intel.patch

2009-03-12 Thread Paul Menzel
Am Donnerstag, den 12.03.2009, 19:08 +0100 schrieb Thomas Hilber:
> On Thu, Mar 12, 2009 at 06:31:47PM +0100, Thomas Hilber wrote:
> > > a) Why is MIN_CLOCK set to 25000 in intelfbhw.h [3]? What would be the
> > > downside of setting it to 1?
> > 
> > MIN_CLOCK is set to 12000 by the patch. What allows for SCART suitable
> > dotclocks. I don't know why the original driver denies such low
> > clock frequencies.
> 
> I forgot to say patching the intelfb is only needed if you even want to
> run the linux console in VGA2SCART mode.
> 
> For the Xserver in VGA2SCART+FRC mode it's not needed.

Thanks. I figured that after reading a lot of threads. (And thanks for
your other answer.)

I hope, you have seen my message to linux-fbdev-devel [1]. I will try to
do what Krzysztof suggested, but will have to read more about that.
Maybe we get it into Linux kernel 2.6.30.

Should we continue discussion about this on linux-fbdev-devel, that
means, can subscribe there?


Thanks,

Paul


[1] 
http://sourceforge.net/mailarchive/forum.php?thread_name=1236701836.4068.85.camel%40mattotaupa.wohnung.familie-menzel.net&forum_name=linux-fbdev-devel


signature.asc
Description: Dies ist ein digital signierter Nachrichtenteil
___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


[vdr] Popcorn as VDR frontend

2009-03-12 Thread Johan Andersson
I need some advise on routes to look into on the subject.

First, thank you Klaus and all for a great PVR. I have tried a lot of 
others like Myth, MediaPortal, GBPVR etc, but the silent menusystems are 
hopelessly inferior to having all menus on OSD. Me and my family have 
used VDR for 2+ years now and we are all happy with it.

I am not bleeding edge VDR with DVB-T in Sweden being SDTV and we have 
only one TV-set although I'm running client/server xineliboutpout with a 
  dedicated frontend machine.

I bought myself a PCH with the idea of using it instead of my PC VDR 
frontend.

With some effort and googling I got xineliboutputs vdr-fbfe (with 
xinelib et.al) cross compiled and runnable on the PCH.

However AFAIK, the directfb usage in xinelib is based on pixelmaps 
written to a dfb surface, all mpeg decode is done in software, this is 
suboptimal on the PCH along with sound being a problem. Running vdr-fbfe 
on PCH says 'DFBGetSurface() not supported' and exits.

It seems people have looked into streamdev instead. I like my VDR OSD 
though on the TV-set, am I wrong in believing that is a nono with streamdev?

Now PCH has an 'IAdvancedMediaProvider' extension to directfb seemingly 
allowing the mpeg-stream to be sent directly to the hardware, thereby 
including both video and audio, it is sadly not fully documented in the 
headers released by Syabas...

The most promising option as I see it is to slash down vdr-fbfe to only 
display the 'osd_command' data on a PCH DFB layer without video 
capability and stream 'normally' to PCH using ideally the xineliboutputs 
servers ability to supply for example an http stream. The hope being 
that the 'vdr-fbfe' application can access directfb concurrently and 
have the OSD displayed ontop of the video layer.

Before I dig deeper into this I would like to ask the community what has 
been tried and what routes do you see as most viable?

I know of some other PCH frontend software for GBPVR Myth, ie vomp and 
mvcpmx(?) but they seem to return me to the world of silent menus, not a 
place I want to be.

/Johan


-- 
Johan Andersson, j...@jna.pp.se


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] [VGA sync field] question to fb-radeon-intel.patch

2009-03-12 Thread Thomas Hilber
On Thu, Mar 12, 2009 at 06:31:47PM +0100, Thomas Hilber wrote:
> > a) Why is MIN_CLOCK set to 25000 in intelfbhw.h [3]? What would be the
> > downside of setting it to 1?
> 
> MIN_CLOCK is set to 12000 by the patch. What allows for SCART suitable
> dotclocks. I don't know why the original driver denies such low
> clock frequencies.

I forgot to say patching the intelfb is only needed if you even want to
run the linux console in VGA2SCART mode.

For the Xserver in VGA2SCART+FRC mode it's not needed.

- Thomas

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] vdr 1.7.4 HD-Recording not working

2009-03-12 Thread Oliver Bardenheier
I tried a complete vanilla VDR from scratch, without any output device (NO 
xine) or any plugins
latest hg-v4l-dvb drivers
same effect:  (ok here we see a timeout)

Mar 12 17:15:43 Coruscant vdr: [31690] timer 21 (36 1650-1730 'X:enius') set to 
event Don 12.03.2009 16:55-17:25 (VPS: 12.03 16:54) 'X:enius'   
   
Mar 12 17:15:43 Coruscant vdr: [31690] switching device 1 to channel 36 

   
Mar 12 17:15:43 Coruscant vdr: [31690] timer 21 (36 1650-1730 'X:enius') start  

   
Mar 12 17:15:43 Coruscant vdr: [31690] Title: 'X:enius' Subtitle: 
'Wissensmagazin: Walforschung'  
 
Mar 12 17:15:43 Coruscant vdr: [31690] record 
/video/X:enius/2009-03-12.16.50.36-0.rec
 
Mar 12 17:15:43 Coruscant vdr: [31690] cFileName::SetOffset: removing 
zero-sized file /video/X:enius/2009-03-12.16.50.36-0.rec/1.ts   
 
Mar 12 17:15:43 Coruscant vdr: [31690] recording to 
'/video/X:enius/2009-03-12.16.50.36-0.rec/1.ts' 
   
Mar 12 17:15:43 Coruscant vdr: [31734] recording thread started (pid=31690, 
tid=31734)  
   
Mar 12 17:15:43 Coruscant vdr: [31733] TS buffer on device 1 thread ended 
(pid=31690, tid=31733)  
 
Mar 12 17:15:43 Coruscant vdr: [31732] buffer stats: 119004 (5%) used   

   
Mar 12 17:15:43 Coruscant vdr: [31732] receiver on device 1 thread ended 
(pid=31690, tid=31732)  
  
Mar 12 17:15:43 Coruscant vdr: [31735] receiver on device 1 thread started 
(pid=31690, tid=31735)  

Mar 12 17:15:43 Coruscant vdr: [31736] TS buffer on device 1 thread started 
(pid=31690, tid=31736)  
   
Mar 12 17:15:44 Coruscant kernel: stb6100_set_bandwidth: Bandwidth=47125000 

   
Mar 12 17:15:44 Coruscant kernel: stb6100_get_bandwidth: Bandwidth=4800 

   
Mar 12 17:15:44 Coruscant kernel: stb6100_set_frequency: Frequency=1611000  

   
Mar 12 17:15:44 Coruscant kernel: stb6100_get_frequency: Frequency=1610982  

   
Mar 12 17:15:44 Coruscant kernel: stb6100_set_bandwidth: Bandwidth=47125000 

   
Mar 12 17:15:44 Coruscant kernel: stb6100_get_bandwidth: Bandwidth=4800 

   
Mar 12 17:15:45 Coruscant kernel: stb6100_set_frequency: Frequency=1611000  

   
Mar 12 17:15:45 Coruscant kernel: stb6100_get_frequency: Frequency=1610982  

   
Mar 12 17:15:45 Coruscant kernel: stb6100_set_bandwidth: Bandwidth=47125000 

   
Mar 12 17:15:45 Coruscant kernel: stb6100_get_bandwidth: Bandwidth=4800 

   
Mar 12 17:15:45 Coruscant kernel: stb6100_set_frequency: Frequency=1611000  

  

Re: [vdr] [VGA sync field] question to fb-radeon-intel.patch

2009-03-12 Thread Thomas Hilber
On Wed, Mar 11, 2009 at 06:15:04PM +0100, Paul Menzel wrote:
> 1. In your patchset the patch fb-radeon-intel.patch [1] does include
> changes to drivers/video/intelfb/intelfbhw.c although in the install
> instructions [2] it is stated that intel drivers do not need to be
> modified.

sorry, install instructions are not up to date for the Intel version
of the patch.

> 
> The changelog in 0.11 includes these two items.
> 
> - patch against intelfb (kernel 2.6.26) to allow for PAL/SCART
>   video timings. You now can use a regular SCART CRT as display
>   for linux console.
> - fixed a bug in intelfb initialization which sporadically
>   setup video timing with weirdous values
> 
> I guess the installation instructions need to be updated for release
> 0.11.

right. as said installation instructions have not been touched since
a while. I just included the plain Intel patches into the package.
There was not yet too much interest in the patches anyway. So I 
did not want to waste my time for documentation:-) 

> If this is correct I would update the instructions.

thank you for that.

> a) Why is MIN_CLOCK set to 25000 in intelfbhw.h [3]? What would be the
> downside of setting it to 1?

MIN_CLOCK is set to 12000 by the patch. What allows for SCART suitable
dotclocks. I don't know why the original driver denies such low
clock frequencies.

> b) Where do I get information about the registers, like what DPLL_A and
> DPLL_VCO_ENABLE is? In the header file some values are assigned to them,
> but where is it documented what they mean?

DPLL_A (DPLLA_CTRL-DPLL A Control Register) is documented here:

IntelĀ® 965 Express Chipset Family, Volume Three
Section: Display Clock Control Registers

The doc can be found here:
http://intellinuxgraphics.org/documentation.html

Cheers
  Thomas


___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr


Re: [vdr] Pvrinput and HVR1300

2009-03-12 Thread Kartsa
Martin Dauskardt kirjoitti:
> try www.amazon.de
> They sell PVR 150 for 77,98 Euro
> It seems they deliver to every european country.
> http://www.amazon.de/gp/help/customer/display.html?nodeId=504950#ausland
>   
I found that Hauppauge sells in their own webshop PVR250 for 42.88 which 
is by far the cheapest I could find :) They have 60% off. Seems its a 
product they want to get rid of.

___
vdr mailing list
vdr@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr