I just changed the CPU in my system from a 1.60a to a 2.66. No other
changes.

There is now a 3 minute (literally) pause in XP bootup. It occurs after
the initial XP boot screen, but before the desktop appears. The screen
is completely black and the system is totally non-responsive (including
caps lock and num lock keys on the keyboard). But it does pass, in 3
minutes, and the system boots and operates normally thereafter.

This pause does not occur if I slow the CPU FSB down to 100/500 instead
of 133/533 (doing so gives a 2.0 GHz CPU).

The pause does not occur in Windows 98SE (the system is dual boot).

Anyone have any idea what's going on?

The system is an Asus P4T533 with 512 meg of PC1066 RDRAM and a 200 gig
hard drive (WD2000JB), ATI AIW Radeon 8500.

Not the worst thing that could happen, but 3 minutes is a LONG time.

Report this post to a moderator

I wonder if that might have something to do with XP
deciding if the change is small enough to let you
continue using it. That sounds like about the right
time for it to be checking for such a thing.

Just a thought.

Cheers
--

Nick
A+, Network+, MCP

Report this post to a moderator

Almost sounds like a timing problem - in the hardware.
Years ago when I designed processor cards, I remember the ready line (not
ready line)
was always very capable of locking up the system on a marginal/incorrect
logic!
The DMA req line could do the same thing.
Not sure why Win98 would work, except that maybe twice as many data lines
may be going to 0 or 1 with
the XP 32 bit operations and the possible noise implications.
Those effects were always fun to counter.
Would be nice if you could look at one of the R/W lines to see if there is
ANYTHING going on when that
happens. Maybe the heatup/cooldown changes the timing just enough for it to
recover.
Just a guess...
J
--
Check my web site for tips on homenetworking and safe computing
www.pccitizen.com

"Barry Watzman" <Watzman@neo.rr.com> wrote in message
news:HhA_a.84629$ib2.20526404@twister.neo.rr.com...
> I just changed the CPU in my system from a 1.60a to a 2.66. No other
> changes.
>
> There is now a 3 minute (literally) pause in XP bootup. It occurs after
> the initial XP boot screen, but before the desktop appears. The screen
> is completely black and the system is totally non-responsive (including
> caps lock and num lock keys on the keyboard). But it does pass, in 3
> minutes, and the system boots and operates normally thereafter.
>
> This pause does not occur if I slow the CPU FSB down to 100/500 instead
> of 133/533 (doing so gives a 2.0 GHz CPU).
>
> The pause does not occur in Windows 98SE (the system is dual boot).
>
> Anyone have any idea what's going on?
>
> The system is an Asus P4T533 with 512 meg of PC1066 RDRAM and a 200 gig
> hard drive (WD2000JB), ATI AIW Radeon 8500.
>
> Not the worst thing that could happen, but 3 minutes is a LONG time.
>

Report this post to a moderator

No, it's not a product activation issue. Of that I am sure. The CPU
has changed, which XP knows (and allows), but XP does not measure the
speed, and at 100/400 the resulting CPU speed of 2 GHz is still a lot
different from the old speed of 1.6 GHz. Not only that, but product
activation doesn't take any time to "think about" whether reactivation
is required or not.

I don't know what it is, but that's not it.


Nick wrote:

> I wonder if that might have something to do with XP deciding if the
> change is small enough to let you continue using it. That sounds like
> about the right time for it to be checking for such a thing.
>
> Just a thought.
>
> Cheers

Report this post to a moderator

I don't buy that either. Windows 98 works fine, and XP works fine,
there is just the damn 3 minute pause in the boot process for XP only,
with a 533 MHz FSB only. The pause seems to occur right after the
system loads AGP440.SYS, based on the boot logs. The pause is even
present in "safe mode".

On the one hand, I'd make a wager that if I did a clean install, the
pause would go away. On the other hand, I'm really loathe to do that,
this system is working great, a clean install would take 3-4 days, and
I'd lose some settings and configuration that I'd probably never get back.

Now I'm going to scare you: This installation is about 18 months old.
It's on it's 3rd motherboard, it's 3rd video card, it's 3rd CPU, 2nd
sound card and it's 2nd hard drive (which crossed the 137 gig barrier)
without being redone from scratch. With tons of software installs,
upgrades and removals. And it's rock-solid stable, runs months between
lockups, so there has been no reason to do a complete reinstall. And
there probably isn't now, but the 3 full minute pause is annoying,
although XP is fine when it does finally comes up.

For anyone wondering about Product Activation (PA), if you keep the same
network card, and change only one thing at a time and not everything at
once, PA isn't much of an issue. You get 3 changes if the network card
is one of them, or up to 6 if the network card is unchanged, and the
entire database resets completely every 4 months. If I'd changed
everything at once, I would have had a PA issue. But I only changed one
thing at a time, usually more than 4 months apart, and I never
reformatted the hard drive except on the one occasion when it was
changed (PA looks at both the hard drive volser and the ID string
returned by the "identify drive" command, which in most cases includes
the mfgrs. serial number of the drive). I also kept the network card
the same (the NIC card MAC address is a "super weighted" parameter in
the PA scheme), and PA hasn't been an issue.

Although I'd like to get rid of the bootup pause, I'm not sure it's
worth a clean install. And the stability of this system -- that
includes XP and the motherboard / chipset (850e) / memory (Rambus)
combination, is incredible.



John D Loop wrote:

> Almost sounds like a timing problem - in the hardware.
> Years ago when I designed processor cards, I remember the ready line (not
> ready line)
> was always very capable of locking up the system on a marginal/incorrect
> logic!
> The DMA req line could do the same thing.
> Not sure why Win98 would work, except that maybe twice as many data lines
> may be going to 0 or 1 with
> the XP 32 bit operations and the possible noise implications.
> Those effects were always fun to counter.
> Would be nice if you could look at one of the R/W lines to see if there is
> ANYTHING going on when that
> happens. Maybe the heatup/cooldown changes the timing just enough for it to
> recover.
> Just a guess...
> J

Report this post to a moderator

I wouldn't bet on it Barry. I have that problem on one of my old test
boxes. I am certain it has something to do with the Gforce card as it
doesn't occur when I put an old PCI card in.

Report this post to a moderator

In article <3F3ADA68.8020501@neo.rr.com>, Barry Watzman
<Watzman@neo.rr.com> wrote:

> I don't buy that either. Windows 98 works fine, and XP works fine,
> there is just the damn 3 minute pause in the boot process for XP only,
> with a 533 MHz FSB only. The pause seems to occur right after the
> system loads AGP440.SYS, based on the boot logs. The pause is even
> present in "safe mode".
>
> On the one hand, I'd make a wager that if I did a clean install, the
> pause would go away. On the other hand, I'm really loathe to do that,
> this system is working great, a clean install would take 3-4 days, and
> I'd lose some settings and configuration that I'd probably never get back.
>
> Now I'm going to scare you: This installation is about 18 months old.
> It's on it's 3rd motherboard, it's 3rd video card, it's 3rd CPU, 2nd
> sound card and it's 2nd hard drive (which crossed the 137 gig barrier)
> without being redone from scratch. With tons of software installs,
> upgrades and removals. And it's rock-solid stable, runs months between
> lockups, so there has been no reason to do a complete reinstall. And
> there probably isn't now, but the 3 full minute pause is annoying,
> although XP is fine when it does finally comes up.
>
> For anyone wondering about Product Activation (PA), if you keep the same
> network card, and change only one thing at a time and not everything at
> once, PA isn't much of an issue. You get 3 changes if the network card
> is one of them, or up to 6 if the network card is unchanged, and the
> entire database resets completely every 4 months. If I'd changed
> everything at once, I would have had a PA issue. But I only changed one
> thing at a time, usually more than 4 months apart, and I never
> reformatted the hard drive except on the one occasion when it was
> changed (PA looks at both the hard drive volser and the ID string
> returned by the "identify drive" command, which in most cases includes
> the mfgrs. serial number of the drive). I also kept the network card
> the same (the NIC card MAC address is a "super weighted" parameter in
> the PA scheme), and PA hasn't been an issue.
>
> Although I'd like to get rid of the bootup pause, I'm not sure it's
> worth a clean install. And the stability of this system -- that
> includes XP and the motherboard / chipset (850e) / memory (Rambus)
> combination, is incredible.
>
>

You seem to have access to alot of equipment... try replacing this drive
with another one to test your theory... Just put in a small HDD and load
a clean install on it... see what happens... If it doesnt solve
anything, you haven't lost anything- just put the original HDD back in
place.

Report this post to a moderator

For the benefit of anyone else who runs into this problem with the
P4T533 -- the problem is a LONG delay during bootup of Windows XP -- I
have found the cause of the problem, it's a BIOS bug that is present in
the 1006 BIOS for the P4T533, but which is not present in the 1005 BIOS.
I have temporarily reverted to the 1005 BIOS.

Since I posted this, I've been contacted by other people who have run
into this problem on other Asus motherboards. That's not unlikely,
since I'm sure that lots of the BIOS code is common across various
motherboards. In one case, the hangup was a full 15 minutes with
Windows 2000 and certain combinations of CPUs.


Barry Watzman wrote:

> I just changed the CPU in my system from a 1.60a to a 2.66. No other
> changes.
>
> There is now a 3 minute (literally) pause in XP bootup. It occurs after
> the initial XP boot screen, but before the desktop appears. The screen
> is completely black and the system is totally non-responsive (including
> caps lock and num lock keys on the keyboard). But it does pass, in 3
> minutes, and the system boots and operates normally thereafter.
>
> This pause does not occur if I slow the CPU FSB down to 100/500 instead
> of 133/533 (doing so gives a 2.0 GHz CPU).
>
> The pause does not occur in Windows 98SE (the system is dual boot).
>
> Anyone have any idea what's going on?
>
> The system is an Asus P4T533 with 512 meg of PC1066 RDRAM and a 200 gig
> hard drive (WD2000JB), ATI AIW Radeon 8500.
>
> Not the worst thing that could happen, but 3 minutes is a LONG time.
>

Report this post to a moderator

Glad you found the answer, I was sure you would find it and post it.

"Barry Watzman" <Watzman@neo.rr.com> wrote in message
news:Obz%a.105667$ib2.22524545@twister.neo.rr.com...
> For the benefit of anyone else who runs into this problem with the
> P4T533 -- the problem is a LONG delay during bootup of Windows XP -- I
> have found the cause of the problem, it's a BIOS bug that is present in
> the 1006 BIOS for the P4T533, but which is not present in the 1005 BIOS.
> I have temporarily reverted to the 1005 BIOS.
>
> Since I posted this, I've been contacted by other people who have run
> into this problem on other Asus motherboards. That's not unlikely,
> since I'm sure that lots of the BIOS code is common across various
> motherboards. In one case, the hangup was a full 15 minutes with
> Windows 2000 and certain combinations of CPUs.
>
>
> Barry Watzman wrote:
>
> > I just changed the CPU in my system from a 1.60a to a 2.66. No other
> > changes.
> >
> > There is now a 3 minute (literally) pause in XP bootup. It occurs after
> > the initial XP boot screen, but before the desktop appears. The screen
> > is completely black and the system is totally non-responsive (including
> > caps lock and num lock keys on the keyboard). But it does pass, in 3
> > minutes, and the system boots and operates normally thereafter.
> >
> > This pause does not occur if I slow the CPU FSB down to 100/500 instead
> > of 133/533 (doing so gives a 2.0 GHz CPU).
> >
> > The pause does not occur in Windows 98SE (the system is dual boot).
> >
> > Anyone have any idea what's going on?
> >
> > The system is an Asus P4T533 with 512 meg of PC1066 RDRAM and a 200 gig
> > hard drive (WD2000JB), ATI AIW Radeon 8500.
> >
> > Not the worst thing that could happen, but 3 minutes is a LONG time.
> >
>

Report this post to a moderator

Well, I did find it, but it wasn't what I posted. That solution proved
to be a phantom.

The problem is an ATechFlash Pro II 6-in-1 internal USB flash card reader.

For some reason, when it is plugged into a motherboard (Asus P4T533)USB
1.1 port, and when the CPU has a 533 MHz FSB, it causes Windows XP to
pause for 3 minutes during the boot process. This does not happen under
Windows 98SE, and it does not happen under XP if the CPU FSB is 400 MHz.
The USB port is working properly, if you sit out the 3 minute delay
(plus the normal boot time), when the system does boot, the flash card
reader works. This even happens with a clean install of Windows XP and
NO other software on a spare hard drive.

The work around (I won't call it a fix) was to put the flash reader
(which is a USB 1.1 device) on one of the USB 2.0 ports. It's a waste
of a USB 2.0 port, but otherwise it solves all aspects of the problem.

Now I know of absolutely no mechanism that would explain WHY this is
happening, because it's sensitive to the CPU FSB, yet USB is an
asynchronous serial bus.

Go figure. At least I found out what was happening and a workaround.

[by the way, I have lots of other USB devices connected to this computer
without problems of any type.]



Pikoro wrote:

> Glad you found the answer, I was sure you would find it and post it.
>
> "Barry Watzman" <Watzman@neo.rr.com> wrote in message
> news:Obz%a.105667$ib2.22524545@twister.neo.rr.com...
>
>>For the benefit of anyone else who runs into this problem with the
>>P4T533 -- the problem is a LONG delay during bootup of Windows XP -- I
>>have found the cause of the problem, it's a BIOS bug that is present in
>>the 1006 BIOS for the P4T533, but which is not present in the 1005 BIOS.
>> I have temporarily reverted to the 1005 BIOS.
>>
>>Since I posted this, I've been contacted by other people who have run
>>into this problem on other Asus motherboards. That's not unlikely,
>>since I'm sure that lots of the BIOS code is common across various
>>motherboards. In one case, the hangup was a full 15 minutes with
>>Windows 2000 and certain combinations of CPUs.
>>
>>
>>Barry Watzman wrote:
>>
>>
>>>I just changed the CPU in my system from a 1.60a to a 2.66. No other
>>>changes.
>>>
>>>There is now a 3 minute (literally) pause in XP bootup. It occurs after
>>>the initial XP boot screen, but before the desktop appears. The screen
>>>is completely black and the system is totally non-responsive (including
>>>caps lock and num lock keys on the keyboard). But it does pass, in 3
>>>minutes, and the system boots and operates normally thereafter.
>>>
>>>This pause does not occur if I slow the CPU FSB down to 100/500 instead
>>>of 133/533 (doing so gives a 2.0 GHz CPU).
>>>
>>>The pause does not occur in Windows 98SE (the system is dual boot).
>>>
>>>Anyone have any idea what's going on?
>>>
>>>The system is an Asus P4T533 with 512 meg of PC1066 RDRAM and a 200 gig
>>>hard drive (WD2000JB), ATI AIW Radeon 8500.
>>>
>>>Not the worst thing that could happen, but 3 minutes is a LONG time.
>>>
>>
>
>

Report this post to a moderator