i965G (Asus P5B-V): Diagnosing ADD2/SDVO issues
reenoo at gmx.de
Thu Mar 1 10:54:42 PST 2007
On Thu, 2007-03-01 at 19:50 +0100, Rene Wagner wrote:
> Hi Keith,
> On Wed, 2007-02-28 at 09:49 -0800, Keith Packard wrote:
> > On Wed, 2007-02-28 at 16:23 +0100, Rene Wagner wrote:
> > > Now, I get a signal on the DVI port for a few seconds after switching
> > > the computer on. Then it just dies. The intel driver shows that an
> > > SDVO adapter is present in the register dump, but then fails to
> > > recognize a device (log using a git build as of last night attached).
> > The driver sees the SDVO channel on the chip as it does with all
> > hardware that has SDVO, but it can't detect anything on the line with
> > I2C, which seems to strongly indicate that the card and the motherboard
> > aren't talking.
> OK, thanks for confirming.
> > That the BIOS also fails to initialize the SDVO link is another
> > indication that the card isn't compatible with the motherboard; the BIOS
> > should auto-detect and configure the card.
> Would there be any DVI signal before the BIOS does the configuration?
> The symptom I see is that I do sometimes get a signal up until grub
> > > Does anyone have any recommendations how to diagnose the problem? Is
> > > this a case of incompatible hardware, a broken BIOS, or is the hardware
> > > just faulty? Reports of any other ADD2/SDVO cards working with the
> > > Asus P5B-V board would also be appreciated.
> > I haven't tried this board; do you know if it uses ADD2-R cards instead
> > of the ADD2-N card?
> Asus sells an ADD2-R just for one particular board, the N4L-VM DH. For
> all other boards with i915, i945, i965 chipsets they sell an ADD2-N
> The Asus manual that's now linked to from the X.org Wiki (thanks guys!)
> indicates that it's also a Silicon Image card, but based on the
> Sil 1632 whereas the card I have comes with an Sil 1642.
That should have read "whereas the card I have comes with an Sil 1364".
> Could that make the difference? I could order the Asus card once any
> reseller has it in stock again.
> The other thing I found at  is:
> "The eventual critical step appeared to be to load the Intel driver
> under windows. After this, I was able to get BIOS, linux console, and X
> on the DVI output."
> Could there be any truth in that? It would be the first time any of my
> personal workstations run Windows in five years, but hey...
>  http://lwn.net/Articles/194763/
reenoo at gmx dot de
4F33 7FD7 93B3 166B BADA D6F8 71A1 FEA8 58B4 36D0
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 189 bytes
Desc: This is a digitally signed message part
More information about the xorg