[mythtv-users] More XvMC data points

[Michael Tiller]

On 12/1/06, Patrick Ouellette <pat at flying-gecko.net> wrote:
>
> On Fri, Dec 01, 2006 at 12:07:11PM -0500, Michael Tiller wrote:
> >
> > built MythTV from SVN recently and specifically looked and it said
> something
> > like "OpenGL Sync: Disabled" or some such thing.  That is the correct
> > setting right?  (i.e. the default setting)
>
> I was told to try turning it on and it seems to work for me.


I'm just starting wonder how many of these claims have actually been
substantiated. :-)  So many XvMC related issues seem to be "try standing on
your head and rubbing your stomach".  I'm always left wondering "Are you
sure this makes any difference?" :-)

> using the latest NVidia driver
> >
> > I've got a 440 MX using the 8776 driver.
>
> 9629 driver.  Check the AGP settings in the xorg.conf to be sure you age
> really getting full speed out of the AGP.


It's not an AGP card.

Partially, if you enable the OpenGL vblank at compile time, you get a
> check box in the TV Playback setup screen too (same screen that sets
> "use video as timebase" I think).


I get that setting even without enable OpenGL vsync.

> soundcard buffer settings
> >
> >
> > By this do you mean turn on extra audio buffering during playback? (in
> the
> > Playback settings on the front end)
>
> You can try it - for my system it helps to have it turned on with the
> 9629 video driver, but off with the 8776 driver (don't know why).
>
> I will say XvMC output really, really, really, really sucked on my
> system before 8776, and 9629 is even better.


I'll have to try 9629 just to cover all the bases but I'm not optimistic.

> and dot clock in my modeline have significantly improved my XvMC output.
> >
> >
> > I'm lost here.  I've tried adding my own modelines but it always
> complains
> > they aren't valid




There is a modeline calculator at
> http://xtiming.sourceforge.net/cgi-bin/xtiming.pl
>
> Something around this should be close:
>
> Modeline "720x480 at 60i" 12.60 720 752 792 824 480 491 494 505 interlace


I still get:

(II) NVIDIA(0): Assigned Display Device: TV-0
(WW) NVIDIA(0): No valid modes for "1024x768_60"; removing.
(WW) NVIDIA(0): No valid modes for "720x480 at 60i"; removing.
(WW) NVIDIA(0): No valid modes for "720x480_60"; removing.
(WW) NVIDIA(0): No valid modes for "NTSC_60i"; removing.
(II) NVIDIA(0): Validated modes:
(II) NVIDIA(0):     "1024x768"
(II) NVIDIA(0):     "720x480"
(II) NVIDIA(0): Virtual screen size determined to be 1024 x 768

I'm putting the Modeline here:

Section "Monitor"
>         Identifier   "Monitor0"
>         VendorName   "Monitor Vendor"
>         ModelName    "27in Analog TV"
>         HorizSync    30.0 - 50.0
>         #VertRefresh  60.0 - 60.0
>         Option       "dpms"
>         Option       "UseEdidDpi" "FALSE"
>         Option       "DPI" "100 x 100"
>         ModeLine "NTSC_60i" 14.318 768 784 848 910 483 484 492 525
> Interlace
>         Modeline "720x480_60" 26.72  720 736 808 896  480 481 484 497
> -HSync +Vsync
>         Modeline "720x480 at 60i" 12.60 720 752 792 824 480 491 494 505
> interlace
> EndSection
>

...is this the right place.  I notice that it names the device "TV-0".  Do I
have to use the same name for my monitor?!?



The calculator should be able to come up with something that works.  The
> NVidia
> driver restricts modelines depending on the output format and tv
> standard.
>
> option "TVStandard" "NTSC-M"
> Option "TVOutFormat" "SVIDEO"


Exactly what I have.

or change SVIDEO to COMPOSITE for composite out.
>
> If you are using an external converter, you should be able to get a
> wider variety of modelines to work.


I'm thinking I'm doing something fundamentally wrong since no Modeline I put
in the file seems to work?!?

DISCALIMER: I run the SVN-HEAD version of Myth, YMMV for release
> versions.


I'm running the fixes branch from SVN.

Pat


Thanks for all the help.

--
Mike
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mythtv.org/pipermail/mythtv-users/attachments/20061201/64ec3ca4/attachment.htm