On 03/09/06, Marko Mäkelä <[EMAIL PROTECTED]> wrote: > On Tue, Jul 25, 2006 at 10:23:15AM +0100, Alasdair Campbell wrote: > > Here is a clip of the BBC rolling news channel, I can see LOTs of > > artifacts with the camera movements, and the scrolling text at the > > bottom looks poor: > > > > http://ginezbukov.com/artifacts.tar.bz2 > > I don't know if your sample is still available, but I saw similar artifacts > (16:9 material vertically scaled on 4:3 screen on -vo dfb:mgatv) earlier > this week. I made a short clip that contains some horizontal movement.
Clip is still available if anyone is interested in trying it out - BBC News 24 clip broadcast on UK DVB-T transmission. > Are there any news on the DirectFB front? I would appreciate an > interrupt-driven interface, to avoid burning CPU in the Flip() method. > Besides, the code from DirectFB CVS for better interlaced output broke > the OSD. Has the situation improved since I last checked (June perhaps)? I've been meaning to do more investigating on this, but things have been tough this past month and my girlfriend is certain that she likes vdr and that she doesn't want me messing about with it's inards! I ended up using the Gentoo ebuilds to build my system from scratch because I thought make things easier for me. However they have been even less reliable than my own compiled setup so I'm going back to running CVS versions etc. I'd like to investigate the output using "softdevice -vo dfb:" as I seem to rememebr that although the OSD was broken and there was a mirroring of picture above and below the 'letter box' there were no artifacts, and there was also an option to use the Fb-Intern(al) de-interlacer. Does that have anything to do with the matrox's Interlaced output(?). Some sort of reply from the softdevice programmers or any one from directFB would be useful in determining what could and couldn't work. -- A l a s d a i r C a m p b e l l r a g a w u @ g m a i l . c o m _______________________________________________ Softdevice-devel mailing list [email protected] https://lists.berlios.de/mailman/listinfo/softdevice-devel
