TV out is broken with VT1621 chip

[Andrew Beresford]

I have a EPIA V-10000 board which uses a trident cyberblade/i1 with a VIA VT1621 to do tv-out over the composite connector.

The trident_drv.so in xserver-xorg-driver-trident is broken and displays noise on the screen (it looks as if the refresh rate is totally wrong or as if it is sending NTSC data when I have actually asked for PAL).

The driver from http://www.blackfiveservices.co.uk/EPIATVOut.shtml gets it right - although you have to get tvtool from the same site and turn on PAL output once the xserver has loaded.

[Melkorainur (melkorainur)]

Created an attachment (id=349)
lspci output

[Melkorainur (melkorainur)]

Created an attachment (id=350)
lspci verbose output

[Melkorainur (melkorainur)]

Created an attachment (id=351)
xorg.conf generated using xorg -configure and updated with tvsignal and default
depth set

[Melkorainur (melkorainur)]

Created an attachment (id=352)
Xorg.0.log from using xorg.conf where tv chipset option was set

[Melkorainur (melkorainur)]

Created an attachment (id=353)
xorg.conf from xorg -configure after modifying to set tv chipset and tv screen

[Melkorainur (melkorainur)]

Created an attachment (id=354)
xorg.0.log when using black5's moded trident driver instead of the 6.7 trident
driver

[Alanh-fairlite (alanh-fairlite)]

You should set "TVSignalMode" to "1" to set PAL.

[John-hl (john-hl)]

I have the same problem here (same hardware, same setup).

In console-mode TV out works nicely but in X there is either noise or a black
screen. It did work around XFree86 4.2.0 or so (with the BlackFive driver).

I have tried variations of the following device settings:

Section "Device"
    Identifier "Trident"
    Driver "trident"
    Option "HWCursor" "on"
    Option "PciRetry" "true"
    Option "TVChipset" "VT1621"
    Option "TVSignal" "1"
    Option "TVSignalMode" "1"
# Option "NoPciBurst" "on"
EndSection

If I enable the TV-Out settings the startup of X takes slightly longer so the
driver seems to atleast try do something.

There may be something else I have missed in the configuration file but these
are the only options I could find documented in the man file.

The X.org version is from RH Fedora core development: 6.7.0-4

[Mikerackhabit (mikerackhabit)]

This bug still exists in 6.9.0.

[Andrew Beresford (beezly)]

I have a EPIA V-10000 board which uses a trident cyberblade/i1 with a VIA VT1621 to do tv-out over the composite connector.

The trident_drv.so in xserver-xorg-driver-trident is broken and displays noise on the screen (it looks as if the refresh rate is totally wrong or as if it is sending NTSC data when I have actually asked for PAL).

The driver from http://www.blackfiveservices.co.uk/EPIATVOut.shtml gets it right - although you have to get tvtool from the same site and turn on PAL output once the xserver has loaded.

[Keith Jordan (keith-jordan)]

TV Out is actually broken farther upstream...

i.e the bug is either in xorg or the kernel...

Despite PAL settings in BIOS and xorg.conf, Ubuntu is still pumping out an NTSC signal - wierd. This is the case for other TV-out chips too.

Will this be picked up, or shall I raise a new bug?

[Matt Zimmerman (mdz)]

If the reporter can confirm that the original problem is that the signal is NTSC rather than PAL as configured, then this bug will suffice

[Andrew Beresford (beezly)]

Matt Zimmerman wrote:
> If the reporter can confirm that the original problem is that the signal is NTSC rather than PAL as configured, then this bug will suffice
>
Sorry, I don't have a tv capable of receiving NTSC.

[Daniel Stone (daniels)]

Sorry about the phenomenal bug spam, guys. Adding xorg-team@ to the QA contact so bugs don't get lost in future.

[Jeremy Sequoia (jeremyhu)]

Do you still have hardware to test this? Is it still an issue?

[penalvch (penalvch)]

Andrew Beresford, could you please execute the following via a terminal as it will automatically gather and attach updated debug information to this report:

apport-collect -p xorg 35794

Please ensure you have xdiagnose installed, and that you click the Yes button for attaching additional debugging information.

[Ajax-a (ajax-a)]

Mass closure: This bug has been untouched for more than six years, and is not
obviously still valid. Please reopen this bug or file a new report if you continue to experience issues with current releases.