From: William B. Lurie on
I find, in my Events Viewer, error 45062 as shown
below:

Event Type: Error
Event Source: ati2mtag
Event Category: CRT
Event ID: 45062
Date: 3/3/2010
Time: 8:29:43 AM
User: N/A
Computer: COMPAQ-2006
Description:
CRT invalid display type
Data:
0000: 00 00 00 00 01 00 5a 00 ......Z.
0008: 2c 00 00 00 06 b0 00 c0 ,....�.�
0010: 00 00 00 00 00 00 00 00 ........
0018: 00 00 00 00 00 00 00 00 ........
0020: 00 00 00 00 00 00 00 00 ........

It's related to the ATI, and it doesn't go away when I
download and reinstall the drivers. Device Manager shows the
correct drivers. The CRT hasn't been changed in any way
since the machine went into service 3 years ago. ?????
From: VanguardLH on
William B. Lurie wrote:

> I find, in my Events Viewer, error 45062 as shown
> below:
>
> Event Type: Error
> Event Source: ati2mtag
> Event Category: CRT
> Event ID: 45062
> Date: 3/3/2010
> Time: 8:29:43 AM
> User: N/A
> Computer: COMPAQ-2006
> Description:
> CRT invalid display type
> Data:
> 0000: 00 00 00 00 01 00 5a 00 ......Z.
> 0008: 2c 00 00 00 06 b0 00 c0 ,....�.�
> 0010: 00 00 00 00 00 00 00 00 ........
> 0018: 00 00 00 00 00 00 00 00 ........
> 0020: 00 00 00 00 00 00 00 00 ........
>
> It's related to the ATI, and it doesn't go away when I
> download and reinstall the drivers. Device Manager shows the
> correct drivers. The CRT hasn't been changed in any way
> since the machine went into service 3 years ago. ?????

ati2mtag: ATI Radeon WindowsNT Miniport Driver

So is there really some problem with your host that you chose to not divulge
here, like it is freezing, rebooting, BSODs, or other video or hardware
problems? ATI's drivers have long been known to spew out irrelevant and
vaporous errors.

Try a different version of the Catalyst driver. I don't remember for which
versions but let's say 7.6-7.8 caused an ati2mtag error. Try versions
before and after that to see if the error goes away (it is a bogus error and
isn't a problem other than users getting upset to see it). When you change
to a different version driver, you need to check its use with your
video-critical applications, like games, which are sensitive to changes in
the video driver. It took me testing through many different versions of the
Catalyst driver to find which was the best (but not perfect) one for use on
my host with my OS and suite of applications and games. The result was that
that best choice for my applications was an older driver version but which
ended up spewing out a bogus error in the event logs. Later I found that
one more minor version earlier didn't have the bogus error and would've
probably been just as compatible. I wasn't interested in eradicating errors
in the event log that had no real effect on the behavior of my host. It had
to do with the driver doing an incorrect detect of a monitor on the
unconnected external port on the video card for dual monitors (your card may
have gone cheap and not provided the 2nd video port but its support was
still implemented in the chipset on the video card).

After finding the best Catalyst driver version for my host and programs, it
was probably a couple years before I decided to test newer versions. At
that time, I moved to 9.3. While the original ati2mtag error disappeared, a
different one showed up (something about VIP). Since everything works
despite the error, I'm not wasting time on solving a problem that exists
only in a bogus error. It's now up to version 10.2 but I don't care.
Everything works so there is no need to change to a different driver to only
encounter new bugs in the new code. Only if I install some program that
forces me to test if a later Catalyst version will fix some problems with
that new software would I bother to change. Regarding drivers, and unless a
newer version actually fixes a problem with your software, newer isn't
better. Instead newer is just different.

If everything is working regarding video operation, just ignore the bogus
error. ATI repeatedly generates versions of Catalyst that produce bogus
errors. Sometimes the older drivers are a better choice.
From: William B. Lurie on
VanguardLH wrote:
> William B. Lurie wrote:
>
>> I find, in my Events Viewer, error 45062 as shown
>> below:
>>
>> Event Type: Error
>> Event Source: ati2mtag
>> Event Category: CRT
>> Event ID: 45062
>> Date: 3/3/2010
>> Time: 8:29:43 AM
>> User: N/A
>> Computer: COMPAQ-2006
>> Description:
>> CRT invalid display type
>> Data:
>> 0000: 00 00 00 00 01 00 5a 00 ......Z.
>> 0008: 2c 00 00 00 06 b0 00 c0 ,....�.�
>> 0010: 00 00 00 00 00 00 00 00 ........
>> 0018: 00 00 00 00 00 00 00 00 ........
>> 0020: 00 00 00 00 00 00 00 00 ........
>>
>> It's related to the ATI, and it doesn't go away when I
>> download and reinstall the drivers. Device Manager shows the
>> correct drivers. The CRT hasn't been changed in any way
>> since the machine went into service 3 years ago. ?????
>
> ati2mtag: ATI Radeon WindowsNT Miniport Driver
>
> So is there really some problem with your host that you chose to not divulge
> here, like it is freezing, rebooting, BSODs, or other video or hardware
> problems? ATI's drivers have long been known to spew out irrelevant and
> vaporous errors.
>
> Try a different version of the Catalyst driver. I don't remember for which
> versions but let's say 7.6-7.8 caused an ati2mtag error. Try versions
> before and after that to see if the error goes away (it is a bogus error and
> isn't a problem other than users getting upset to see it). When you change
> to a different version driver, you need to check its use with your
> video-critical applications, like games, which are sensitive to changes in
> the video driver. It took me testing through many different versions of the
> Catalyst driver to find which was the best (but not perfect) one for use on
> my host with my OS and suite of applications and games. The result was that
> that best choice for my applications was an older driver version but which
> ended up spewing out a bogus error in the event logs. Later I found that
> one more minor version earlier didn't have the bogus error and would've
> probably been just as compatible. I wasn't interested in eradicating errors
> in the event log that had no real effect on the behavior of my host. It had
> to do with the driver doing an incorrect detect of a monitor on the
> unconnected external port on the video card for dual monitors (your card may
> have gone cheap and not provided the 2nd video port but its support was
> still implemented in the chipset on the video card).
>
> After finding the best Catalyst driver version for my host and programs, it
> was probably a couple years before I decided to test newer versions. At
> that time, I moved to 9.3. While the original ati2mtag error disappeared, a
> different one showed up (something about VIP). Since everything works
> despite the error, I'm not wasting time on solving a problem that exists
> only in a bogus error. It's now up to version 10.2 but I don't care.
> Everything works so there is no need to change to a different driver to only
> encounter new bugs in the new code. Only if I install some program that
> forces me to test if a later Catalyst version will fix some problems with
> that new software would I bother to change. Regarding drivers, and unless a
> newer version actually fixes a problem with your software, newer isn't
> better. Instead newer is just different.
>
> If everything is working regarding video operation, just ignore the bogus
> error. ATI repeatedly generates versions of Catalyst that produce bogus
> errors. Sometimes the older drivers are a better choice.
Vang, I have no actual known problem, but when I
see red X's in the Events log, I feel that I have a corrupted
system and I don't like that. This all started because I wanted
to know why, sometimes, the ATI function doesn't function right.
Like, I'll go away and expect the ATI to turn things off and
hibernate in a few hours, and I return 8 hours later and find
Desktop on, not even Screen-Saver. So we zeroed in on ATI and
we're still digging. The answer to your question is, video operation
is working fine, but ATI is questionable.

Getting even a download of the Catalyst drivers now installed is
not easy, and getting multiple ones to try is just going too far,
even tho' it does make sense.