Finally a strange but working solution for "display driver has stoppedworking and has recovered"

[Laszlo Lebrun]

Am 16.01.2013 03:58, schrieb VanguardLH:

Patient: Doctor, it hurts when I do this.
Doctor: Then don't do that.

Doctor: you have experienced several severe heart attacks, just after
starting to climb stairs after a period of inactivity: you should do
some gentle sport regularily.
;-)

 

[Gene Wirchenko]

Am 16.01.2013 03:58, schrieb VanguardLH:

Doctor: you have experienced several severe heart attacks, just after
starting to climb stairs after a period of inactivity: you should do
some gentle sport regularily.
;-)

And what sport is gentler than climbing stairs? <G>

OTOH, if someone just sits around like a blob, he is likely to
become a dead blob.

Sincerely,

Gene Wirchenko

 

[Stan Brown]

It looks like the problem is caused by several things in combination,
and Microsoft points to the video GPU/Driver. On the other hand,
the real issue seems to be related to how the error that causes the time
out to occur is handled. Documents indicate that this process involves
windows system drivers and functions, quite probably Microsoft's
development documentation, and the revision/version levels
of the development software. At the same time, it looks like Microsoft
is saying that the application itself need to be involved with
recovering from the error.

I'm not sure how hey can blame "the application". I get this (when I
do) as part of the login sequence. Of the (few) programs that I have
set to launch automatically, not one tries to display anything on the
screen.

 

[charlie]

I'm not sure how hey can blame "the application"

Evidently, some sort of error handling action by the application is
desired when the app hang occurs. I'd think that there is a OPS video
support design problem that MS didn't figure out how to fix.
Actually, variants of the problem started showing up years ago in XP,
and maybe back to ME.
The commonality from then to now involves very high resolution graphics,
and movement. Two "Games" come to mind, one involving puzzle solving in
a "haunted house" (Like the house of Usher, perhaps,) and the other
later one also involving puzzles and very extensive and detailed
landscapes. (Urhu?) Crysis was another one that often had problems.

 

[Stan Brown]

Evidently, some sort of error handling action by the application is
desired when the app hang occurs.

You quoted me out of context.

What I said, but snipped, is that the only time I ever see this error
is when Windows is logging me in, so there is no "application"
running.

 

[Laszlo Lebrun]

Am 20.01.2013 19:05, schrieb Stan Brown:

You quoted me out of context.

What I said, but snipped, is that the only time I ever see this error
is when Windows is logging me in, so there is no "application"
running.

....if from an application or from a job does not really matter, and you
definitely have some jobs running before the user logging.
Anyhow it seems to become a clear picture, that error is mainly
generated whenever the GPU had nothing to do and just needed to start
drawing a new picture. Now with BOINC configure it to sponsor
permanently 10% of my GPU power to protein research, I just avoid
entering in that situation.
I have got no proof, but i think BOINC is a deamon active before the
login, like the printer sharing daemon does.

 

[charlie]

Am 20.01.2013 19:05, schrieb Stan Brown:
...if from an application or from a job does not really matter, and you
definitely have some jobs running before the user logging.
Anyhow it seems to become a clear picture, that error is mainly
generated whenever the GPU had nothing to do and just needed to start
drawing a new picture. Now with BOINC configure it to sponsor
permanently 10% of my GPU power to protein research, I just avoid
entering in that situation.
I have got no proof, but i think BOINC is a deamon active before the
login, like the printer sharing daemon does.

I don't know who did what to who - - But-
Many of the timeout problems were reduced if not eliminated by changes.
A fair number of the changes were done in app "patches", and in a few
windows patches.

 

[charlie]

I don't know who did what to who - - But-
Many of the timeout problems were reduced if not eliminated by changes.
A fair number of the changes were done in app "patches", and in a few
windows patches.

A big fat "I know nothing" guess.

The win video subsystem is involved, to say the least, and DX also.
I also believe that the software development platform (Configuration?)
is involved, likely due to a lack of correct and complete documentation
by MS.

A fair number of the error details refer to a DX software module.
(Either an MS module or an OEM's tailored dx module.)
Some sort of handshake between the various modules and video drivers
didn't properly happen with an expected result, so a hang occurs,
followed by a timeout, hence the timeout error. The app is usually what
is hung completely. Visible display corruption consistent with corrupted
addresses, video data, and a GPU "crash" is fairly common, and may be
followed by a black or even blue screen. On this P/C, the errors were
originally so bad that Crysis was unplayable, as was Far Cry 2.

With earlier video driver versions, seriously gutting windows logging
and error processing provided some noticeable improvement, and made both
games sort of playable, if you "saved" frequently, about every fifteen
minutes or so. As the video drivers for AMD and NVIDIA GPUs improved,
the crashing and lockup rate dropped quite drastically, extending the
game play to a couple of hours or more. Beyond that time frame, I
usually save and do something else.

 

[Laszlo Lebrun]

Am 23.01.2013 02:52, schrieb charlie:

With earlier video driver versions, seriously gutting windows logging
and error processing provided some noticeable improvement, and made both
games sort of playable, if you "saved" frequently, about every fifteen
minutes or so. As the video drivers for AMD and NVIDIA GPUs improved,
the crashing and lockup rate dropped quite drastically, extending the
game play to a couple of hours or more. Beyond that time frame, I
usually save and do something else.

I think we have different issues. Playing is not my thing. My issues
never happened under load (e.g. video conversion with CUDA support), but
always after a period of idle state (not sleep). I tried EVERY driver
and this had exactly NO influence on the error.

 

[charlie]

Am 23.01.2013 02:52, schrieb charlie:

I think we have different issues. Playing is not my thing. My issues
never happened under load (e.g. video conversion with CUDA support), but
always after a period of idle state (not sleep). I tried EVERY driver
and this had exactly NO influence on the error.

My wife had the problem occur yesterday on the PC she uses to do things
like reading emails from her friends. She doesn't begin to tax the hi
resolution portions of a video card, other than the load caused by the
Aero video features. Also, the video card is one of the older HD series
cards, and uses what are now "legacy" drivers, since the drivers were
separated from the combined driver update package sometime in 2012.

Software and the windows system are relatively stable on her machine,
with only windows updates and updates to firefox, thunderbird, and the
microsoft keyboard and mouse. Video conversion with GPU support sounds
like load to me.