No APM with Pentium4 HT ("multiprocessor") possible at all?

[Torsten Blaeser]

Because I have a lot of critical hardware like Audigy, DVB card and so
on I usually configure my Windows XP Prof PC as "Standard PC" and not
the usual ACPI, by pressing F6 during the setup procedure.

Now I upgraded from a normal prfectly working i845/Pentium4 System to
a brand new i875p/Pentium4 2.4c with Hyper Threading.
I installed Windows XP Prof totally new, and like a thousand times
before, pushed F6 in order to install Windows without ACPI.
But now my PC does not shut down automatically, because there is no
"APM" register in the energy options, where I normally can activate
it! I have read the forums and found the hint to manually install the
"NT APM layer" (or whatever it is called in English: I am a German
User) afterwards by going into device manager und using the new
hardware wizard and add the "NT APM ...". I did that, restartet, but
in device manager the "NT APM" is now signed with an exclamation mark,
which means not working. And it really does not work, when I shut down
windows only the "you can now turn off" screen appears and I have to
manually turn it off by pushing the power button.

As you see, I tried everything, what is mentioned in the newsgroups.

Now I have read that APM does not work with multi processor systems
(what the Pentium's with Hyper Threading are in a way), is that
right????

Does that mean, that it is really not possible at all to use APM with
any of the new Pentium4 (except turning off the HT in bios)?
Which would mean that I am forced to use ACPI in order to make my
Windows shut down automatically?
But I don't want to use ACPI because I only had trouble with it!

Or is APM possible with multiprocessor and am I just doing anything
wrong??

Thanx in advance, Torsten

 

[David Hollway]

Because I have a lot of critical hardware like Audigy, DVB card and so
on I usually configure my Windows XP Prof PC as "Standard PC" and not
the usual ACPI, by pressing F6 during the setup procedure.

Now I upgraded from a normal prfectly working i845/Pentium4 System to
a brand new i875p/Pentium4 2.4c with Hyper Threading.
I installed Windows XP Prof totally new, and like a thousand times
before, pushed F6 in order to install Windows without ACPI.
But now my PC does not shut down automatically, because there is no
"APM" register in the energy options, where I normally can activate
it! I have read the forums and found the hint to manually install the
"NT APM layer" (or whatever it is called in English: I am a German
User) afterwards by going into device manager und using the new
hardware wizard and add the "NT APM ...". I did that, restartet, but
in device manager the "NT APM" is now signed with an exclamation mark,
which means not working. And it really does not work, when I shut down
windows only the "you can now turn off" screen appears and I have to
manually turn it off by pushing the power button.

As you see, I tried everything, what is mentioned in the newsgroups.

Now I have read that APM does not work with multi processor systems
(what the Pentium's with Hyper Threading are in a way), is that
right????

Torsten,

APM works fine with MP systems (and that includes HT) but it depends on the
motherboard having implemented it. As ACPI is now the standard for new
systems, some motherboard BIOSes have removed APM support, because it frees
up space in the BIOS for other things (i.e, you only have a 2- or 4-Mbit
Flash chip for the BIOS, and sometimes space is limited). For example, I
believe that some of the Intel branded motherboards have removed APM
support - check the documentation for your particular board.

Does that mean, that it is really not possible at all to use APM with
any of the new Pentium4 (except turning off the HT in bios)?
Which would mean that I am forced to use ACPI in order to make my
Windows shut down automatically?

If your board doesn't support it, then you have two choices:

1) change boards
2) live without power management

But I don't want to use ACPI because I only had trouble with it!

Actually, you have a 3rd choice - use ACPI. If you have problems with it,
then this is the fault of some of your peripherals, not the ACPI standard
itself, which is actually very robust if correctly implemented. I know from
personal experience that Soundblaster cards' drivers often have problems
with ACPI - for example, I've found that they prevent the S3 Suspend-To-RAM
state from working. If I were you, I would prefer to debug the ACPI problems
rather than use APM, which is a very old and limited mechanism. I know this
isn't the answer you want

What problems exactly are you having with ACPI?

Hope this helps..

 

[Torsten Blaeser]

APM works fine with MP systems (and that includes HT) but it depends on the

motherboard having implemented it.

I appreciate your help, and the last part of the sentance might be
true, but the first obviuosly not: Microsoft Knowledge Base Article -
307525 (http://support.microsoft.com/default.aspx?scid=kb;EN-US;307525)
says:

"If a computer is a multiprocessor system, Windows XP does not install
APM support"

And in a techweb article
(http://www.techweb.com/winmag/win2000/insider/2000/10.htm#BUGS) it
says:

"...it can track down some common reasons for APM failure out of the
box, like multiprocessor computers (since APM does not work on
multiprocessor PCs)"

Seems like it is a fact that MP really does not work with APM, for
whatever reasons possible and I have to face the fact that I have to
use ACPI and hope that this time my Audigy goes well with my DVB
card...
I know that this is not the fault of ACPI if some cards are not
truelly 100% compliant with this standard, but this doesn't help me a
lot, if I own such a card like the Audigy... thx creative

 

[bprior]

David,

Please forgive the intrusion but I am on the verge of
rebuilding a dual processor system, which originally I
could not implement ACPI nor APM using Win2k. And tech
pages on Microsoft's server specifically stated that Win2k
did NOT support power management with multiprocessor
systems.

I have been looking voraciously for some evidence that
power management (APM of ACPI) is suported under XP in
multiprocessor systems. Are you aware of any specific
literature indicating as much? The vagrant peripheral
problems aside of course.....

Thanks for any help.