On Thursday 25 May 2006 11:34, Shawn Holland wrote:
Section "ServerFlags" Option "NoPM" "true" EndSection
**This disables APM in the XServer. Most newer hardware uses acpi instead of APM.**
That was easy! ;-)
* (WW) NVIDIA(0): Not using mode "512x384" (not a valid TV mode) Note: This one *definitely* needs investigating. You also have 77 tried/invalid display mode entries... "(vrefresh out of range)" ... "(hsync out of range)" ... "(bad mode clock/interlace/doublescan)" and "(no mode of this name)"
This caught my eye as well. I didn't understand why it was trying so many modes when the xorg.conf told it what modes to use. And there was only one. Is this just a standard probe detection by xorg? To determin which modes are available on your connected display device?
It wouldn't surprise me if the default behavior was to try an array of modes if probing reveals a configuration that isn't compatible with the hardware. That's my take on it, anyway.
and, finally, this highly targeted and informative warning:
* (WW) NVIDIA(0): WAIT (0, 1, 0x2000, 0x00000494, 0x00000494, 0)
:-)
I wonder if this "wait" is whats causing no display!
This also wouldn't surprise me, assuming it can't identify a valid display mode.
* configure the system to work on the primary display, first, without having the TV connected and plan on 'adding' the TV after the card is configured.
I can get it to work on a monitor by itself no problem.
That's definitely a good sign. What I'm thinking is it might be helpful to pare xorg.conf down and refine it to just the entries needed to support your primary display. Theoretically, it should then be easier to 'add' support for the second display which, in this case, is the TV.
* configure the system to work using the TV as the primary display, first, without having your usual primary display (monitor) connected to the system (i.e. "unplug the monitor") then 'adding' your monitor after the card is configured.
I can certainly try this. However, Using SaX2 defaults to use the vga out. I'm not sure how to tell it to go to composite out.
I definitely saw instructions on this... maybe not with SaX2 but possibly with the nvidia-settings(name?) utility or... maybe it was a 'how-to' I read on the web? In any case, if you pursue this route, regardless of a utility or hand-editing xorg.conf, the end result would provide the parameters needed to 'speak' correctly to the TV. I imagine it would use the same settings to drive that output in the dual (not simultaneous) display arrangement. Side question: Did I interpret what I read about that card correctly? I got the impression it handles the 'splitting' between S-video and composite internally... meaning it doesn't matter if you connect through S-video or composite, the signals are available at both outputs simultaneously?
* proceed as you are, attempting to configure the system with both the TV and your primary display connected. I don't recommend this.* (*)I'm a big fan of reducing problems down to the least common denominator... the least complexity... until I can establish a solid foundation (no 'unknowns' at play) that I can confidently build upon. It's your system, your time & effort and, eventually your how-to :-) so obviously the decision here is your's.
This was the easiest solution. And I also used ssh from my desktop to do most of the xorg.conf changes just becasue its a lot easier to copy / paste!
What I'd really like to see (I even Googled a bit for it) is a copy of a working xorg.conf from a similarly configured system. Without the hardware or past experience you've got, I'm still trying to wrap my head around the number of required 'screen' and 'monitor' and 'mode' entries. ;-)
Where its using the Monitor[0] and I left the monitor vendor setup as the NEC and not the tv which use to detect as VESA. I don't know if thats causing a problem.
Yes, that's what I was looking at. But this ties into my comment, above, about coming to grips with what and how many entries are required to support these displays? Now I'm going for some Tylenol ... ;-) Carl