I have been trying to work out what is the best video timing method to use for mythfrontend. I am guessing that using the right timing will improve the deinterlacing of analogue tv signals. In my case, I'm running an Nvidia gforce 5200 mx agp card, connected via dvi to my plasma screen. DVB signals look great on the screen, but analogue tv is woeful. I have experimented with the various filters in mythfrontend and have settled on bob as the best compromise, however because my monitor only wants to run at 60hz instead of 50hz (i'm in pal land), this makes the osd flicker, and the picture is a bit flickery as well.

Even though I've got "use openGL Vsync" selected in the front end settings, the frontend only ever uses USleep with busy wait timing.

I suspect that I may be able to improve things by changing my Xorg.conf file or my nvidia settings, but I'm not really sure what works when using dvi, as opposed to tv out.

Any clarification of all this would be welcome, I did find a post (below) on the ATRPMS site that mentions the different timing methods, but I've been unable to find out what are the pros and cons of the various types

Quote:
"I compiled Myth last night and the OpenGL sync works great on both of my 
NVidia cards.  I didn't have time to try the ATI card.  The first sync 
method to be available is what will be used:

1.  NVidia sync (4363 driver only)
2.  DRM sync (not sure what this is)
3.  OpenGL sync (but has to be enabled in settings.pro, not sure if it 
works on other cards, will test tonight)
4.  RTC sync
5.  Busy Wait

So basically if it fails on one of the sync methods it will try the next 
one.  Busy Wait should work on all systems.  I will test more tonight.

Axel Thimm wrote:
> Thanks, sounds reasonable. Would you also test it on non-OpenGL
> or non-nVidia systems (that's where my concerns are)? Anyone else?
>

-- 
Roger Little"
_______________________________________________
mythtv-users mailing list
[email protected]
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users

Reply via email to