Perhaps it'll seem trivial but I don't realize how to have this feature
working.
I never found that a driver was backed up so that I could see driver
rollback effectively rolling anything back.

I'd like to learn about this issue: does one need to enable something
special, or explicitely choose to save the older copy of drivers?
I know about uninstalling enite applications, or system restore points only
to overcome driver problems...

thx in advance,
ricardo

"ricardo.l" <ricardol@discussions.microsoft.com> wrote in message
news:3D075F8D-EFC8-493B-B47B-4FBF9FDE2356@microsoft.com...
You don't have to do anything. When drivers are installed the .inf, .pnf, .sys and
(if applicable) .cat are placed in a drive on the system. When you update drivers,
those files are place on the drive as well without touching the previous driver
files. As long as you don't delete the files or they don't get corrupted, clicking
the "Roll back driver" button will uninstall the presently installed driver and
reinstall the driver just previous to it.

You can find the files mentioned above in both:
c:\windows\.inf
and
c:\windows\system32\drivers


--

Brian A. Sesko { MS MVP_Shell/User }
Conflicts start where information lacks.
http://basconotw.mvps.org/

Suggested posting do's/don'ts: http://www.dts-l.org/goodpost.htm
How to ask a question: http://support.microsoft.com/kb/555375

"Brian A." wrote:

Thanks for your feedback.
If I can ask, is there a way to rollback drivers on an unbootable xp sp2
system?
From dump analysis it seems that it is a faulty graphic driver (or as I
suspect, a video card utility which failed to install properly), so if I
could find anything in those directories and exploit it to rollback driver
changes, that would help me out.
Or maybe do I need to enable something in the registry?

Of course same question to try uninstalling a faulty application.

Whatever, It would me useful to know how to work it out, assuming that no
restore points are available.
Thank you again for your explanations,
Ricardo

"Brian A." wrote:
....
I'd like to comment on Daniel Petri MVP article, that you include in your
signature.

He wrote in conclusion:
"...However, saying "Thank you in advance" is not, and most people feel bad
about it because they read it as "I have a very urgent question I want you to
help me with, but I don't really have the time to come back and personally
thank you, because it's MY time I care about, not yours"..."

I do not agree:
when I write it, I mean that I care so much about your answer that I feel to
thank you without waiting for replies, which I could never receive.
And rating replies as soon as one reads them, is another way to confirm that
feeling, besides just restating appreciations.