P3B-F and S370-Slot-1 converter ?

[Roland]

Hello,

I want to upgrade my system. I've got now a P3B-F board with a 750MHz
Pentium 3 CPU.

I've got now a Intel Celeron 1300 MHz Tualatin CPU and a converter-card.
The convertercard is a Gigabyte GA-6R7+. On the card I have a S370 socket
and the whole card can connected to the Slot-1 socket on the P3B-F.

But, it will not work... If I place the card and CPU, my computer does
nothing, only thing I hear is a alarm-sound
I've got revision 1.04 of the P3B-F and I updated the bios to version
1008.004.

I use 133 MHz SD-Ram modules, and a 32 MB AGP videocard. I don't know what's
more imported to know for you to help me.

I hope someone can help me, the convertercard works, it's tested on the same
processor and runs very good. But not on my P3B-F board. I saw on many pages
on internet that poeple use a P3B-F board with the same CPU, but mine is not
working.

Can anyone help me with this?

Thanks in advance!


Roland.

 

[Daniel Mandic]

Hi Roland!



I would suggest a NEW Tualatin Slot-1 Adapter (s370 and s370-II, also
known as FC-PGA-II)





I don´t know if the Gigabyte Adpater can hold FC-PGA2 (Tualatin) Cpu´s.
I don´t think so.




Best Regards,

Daniel Mandic

 

[Roland]

Hello Daniel,

Thanks for your messages!
Strange thing is, the CPU is tested on the card that I have, and it works!
But not on my P3B-F board.

I heard something about a reset of the P3B-F with a jumper, I must try this.
Another question is, must I change something on my motherboard or bios to
get it work?

And, I hear about a Slot-T converter... But I don't know if this works...

Many questions and many options, and if one of these options give an error
of make a failure, the whole PC didn't startup... I can search for another
converter-card, maybe I have more luck with that...

Thanks,

Roland.

 

[Daniel Mandic]

Hello Daniel,

Thanks for your messages!
Strange thing is, the CPU is tested on the card that I have, and it
works! But not on my P3B-F board.

Hmmm... as you said, that you have the Beta 1008.004 installed. The
Tualatin should work.
I guess the Adapter cannot hold Tualatins.

I heard something about a reset of the P3B-F with a jumper, I must
try this. Another question is, must I change something on my
motherboard or bios to get it work?

Clear CMOS Ram. Well, you could try. But first, be sure your
(Gigabyte) adapter can adapt Tualatins.

And, I hear about a Slot-T converter... But I don't know if this
works...

I have the best Tualatin Adapter, IMHO.

Many questions and many options, and if one of these options give an
error of make a failure, the whole PC didn't startup... I can search
for another converter-card, maybe I have more luck with that...

Probably.

I don´t know what´s happening to a FC-PGA2 CPU in a FC-PGA Socket, when
turning on power.

Maybe your 1300 Tualatin is already tuned to work in Standard S370
Sockets!?
You can find on the Internet several Tul-Mods... Even all, I would
suggest you a good Adapter... (depends if your CPU is still working!?).
Even that, with a good adapter you can upgrade to the P3-S Tualatin,
too (available again)!

Well, I would try this 750 P3 once again. Check if 1008.004 is properly
installed and don´t use this adapter with the Tualatin, when not
compatible to.



Best Regards,

Daniel Mandic

 

[Paul]

Hello Daniel,

Thanks for your messages!
Strange thing is, the CPU is tested on the card that I have, and it works!
But not on my P3B-F board.

I heard something about a reset of the P3B-F with a jumper, I must try this.
Another question is, must I change something on my motherboard or bios to
get it work?

And, I hear about a Slot-T converter... But I don't know if this works...

Many questions and many options, and if one of these options give an error
of make a failure, the whole PC didn't startup... I can search for another
converter-card, maybe I have more luck with that...

Thanks,

Roland.

Take a look at Roland's FAQ:

http://homepage.hispeed.ch/rscheidegger/p2b_procupgrade_faq.html

Boards with PCB revision 1.03 or greater, appear to have a voltage
regulator suitable for Tualatin processors. Before revision 1.03,
the voltage regulator goes from 1.8V on upwards. Revision 1.03 or
greater has a voltage regulator that allows operation at 1.5V,
which is the voltage that the Tualatin requests.

To solve this problem on my P2B-S, I actually unsoldered the
CPU Vreg chip and replaced it with a pin-for-pin compatible
chip. By using a 1.5V capable regulator, I was able to use a
cheap Upgradeware Slot-T adapter and a Tualatin. If I had chosen
not to change the regulator, I would have needed a Powerleap PL-iP3/T
(the one with a regulator right on the slocket).

If you look at Roland's list of Tualatin capable voltage regulators,
they tend to have the letter "B" on the part number.

HIP6019BCB
HIP6020ACB
HIP6004CB
HIP6004BCB
US3007CW

That chip will be located near the CPU, and on your board, the
ATX main power connector is right next to it.

In terms of the reference material, this early Asus document lists
revision 1.03 as the minimum revision for the proper voltage
regulator chip.

http://web.archive.org/web/20030411...om.de/support/FAQ/faq086b_CPU_Upgrade_III.htm

The current standing page at Asus, lists revision 1.04. In any case,
looking at the voltage regulator chip itself is the "tie breaker"

http://rma.asus.de/support/FAQ/faq086b_CPU_Upgrade_III.htm

Paul

 

[Roland]

Thanks Paul for the great link to that website, I will look for it, maybe I
find a solution!

Roland.

 

[NewMan]

I upgraded a P2B-F with a PowerLeap a couple fo years back. Same deal,
it simply would not work until I flashed the latest beta BIOS. As soon
as I did that, it ran no problem.

Take a look at your card. Seems to me that some of the cards needed to
be modified, or even the motherboards. Some of the early voltage
regulators had problems delivering the power required.

I think that the other web site was mentioned in the other post will
help.

Good luck, these boards really rock when you get them hot-rodded that
fast.

 

[P2B]

Thanks Paul for the great link to that website, I will look for it, maybe I
find a solution!

Roland.

I think we need to go back over this carefully, because you seem to have
all the prerequisites, yet it doesn't work.

You say you have the latest beta BIOS, that's a requirement.

Your board is revision 1.04, which should have a Tualatin-capable
voltage regulator according to Asus. The fact you get an 'alarm' when
you power the system on is a good sign - the board should shut down
immediately and completely (all fans stopped) if it cannot supply the
requested CPU voltage - but check the numbers on your VRM chip as Paul
suggested, just to be sure.

Your slot adapter appears to support Tualatin processors, since you say
the adapter/CPU combination works on another board.

So it should work - but maybe something broke while you were working on
the upgrade.

Make sure your adapter/CPU combination still works on the other board
you tested.

Make sure your P3B-F system still works with the original P3 750 installed.

If all of the above checks out, but the upgrade still won't work, I
think we have to conclude that the CPU/adapter/motherboard combination
is somehow incompatible.

In that case your best bet would be to try an Upgradeware Slot-T adapter
instead. In my experience, they always work on all Asus 440BX chipset
motherboards.

HTH

P2B

 

[Paul]

I think we need to go back over this carefully, because you seem to have
all the prerequisites, yet it doesn't work.

You say you have the latest beta BIOS, that's a requirement.

Your board is revision 1.04, which should have a Tualatin-capable
voltage regulator according to Asus. The fact you get an 'alarm' when
you power the system on is a good sign - the board should shut down
immediately and completely (all fans stopped) if it cannot supply the
requested CPU voltage - but check the numbers on your VRM chip as Paul
suggested, just to be sure.

Your slot adapter appears to support Tualatin processors, since you say
the adapter/CPU combination works on another board.

So it should work - but maybe something broke while you were working on
the upgrade.

Make sure your adapter/CPU combination still works on the other board
you tested.

Make sure your P3B-F system still works with the original P3 750 installed.

If all of the above checks out, but the upgrade still won't work, I
think we have to conclude that the CPU/adapter/motherboard combination
is somehow incompatible.

In that case your best bet would be to try an Upgradeware Slot-T adapter
instead. In my experience, they always work on all Asus 440BX chipset
motherboards.

HTH

P2B

I guess I should be reading the original posts more carefully

The alarm sound would require a working processor to make that
sound. It could be an overheat alarm, for example. Or maybe
an overvoltage ? Or the BIOS thinks it is an undervoltage ?

AFAIK, there is a piece of hardware for making a tone in the
computer speaker. But it has to be programmed by the CPU in
order to make a sound. To make the "European Police Car" sound
requires changing the programming a couple times a second,
and that means the processor is running.

If we assume for a second, that the motherboard is monitoring the
diode temperature, maybe there is a bad connection or problem
with the CPU diode. It could be the hardware monitor is seeing
an abnormal voltage on the CPU diode, because something isn't
making good connections. It likely is not a loose heatsink,
as you would have had trouble while running the slocket on the
other computer.

Is the sound actually the "European Police Car" sound ?

Paul

 

[Roland]

Hello Paul,

I hear the "European Police Sound", two tones of sounds and one is higher
than the other and it's like a police-car! I must wait sometimes for 10
seconds to hear it, but the PC-speaker is making a police-car sound, that's
right!

You can say that my CPU is working allready?

Maybe you've got more info about this...

Roland.

 

[Roland]

By the way, the chip on the mainboard, near the power-connector, is a
HIP6020ACB chip.

 

[P2B]

I guess I should be reading the original posts more carefully

The alarm sound would require a working processor to make that
sound. It could be an overheat alarm, for example. Or maybe
an overvoltage ? Or the BIOS thinks it is an undervoltage ?

AFAIK, there is a piece of hardware for making a tone in the
computer speaker. But it has to be programmed by the CPU in
order to make a sound. To make the "European Police Car" sound
requires changing the programming a couple times a second,
and that means the processor is running.

Quote from the Winbond datasheet:

"Also W83781D uniquely provides an optional feature: early stage (before
BIOS was load) beep warning. This is to detect if the fatal elements
present --- VcoreA, +3.3V voltage fail, and the system can not be boomed
up."

Clearly it is possible for the board to squawk even with no processor
present. The datasheet does not appear to describe this "early stage
beep warning", but my repair notes for one dead P3B-F here say it made a
siren noise even with no CPU or RAM. I can't verify that as the board
has been raided for parts...

P2B

 

[P2B]

By the way, the chip on the mainboard, near the power-connector, is a
HIP6020ACB chip.

Therefore the voltage regulator is not the problem - HIP6020ACB supports
Vcore from 1.3v to 3.5v.

Again, I suggest you verify the cpu/adapter combination still works on
the other board you tested, and that the P3B-F still works with the
original P3 750 cpu.

P2B

 

[Paul]

Quote from the Winbond datasheet:

"Also W83781D uniquely provides an optional feature: early stage (before
BIOS was load) beep warning. This is to detect if the fatal elements
present --- VcoreA, +3.3V voltage fail, and the system can not be boomed
up."

Clearly it is possible for the board to squawk even with no processor
present. The datasheet does not appear to describe this "early stage
beep warning", but my repair notes for one dead P3B-F here say it made a
siren noise even with no CPU or RAM. I can't verify that as the board
has been raided for parts...

P2B

Interesting. Some motherboards contain duplicate functionality, so
you may find speaker drivers on a couple of chips. Perhaps an ohmmeter
from the BEEP pin to the speaker lead, would indicate whether the
chip actually drives the speaker ? With 48mA drive, Asus might consider
connecting that to the speaker directly.

Also, the Winbond datasheet you mention above, has a VID table, and
1.8V is as low as is defined in the table. And, as with all of the
great Winbond datasheets out there, the behavior is not explained in
enough detail to be of any help :-(

Getting the police siren, with no CPU installed, pretty well cements
the case for the W83781D doing this function on its own.

I wonder if trying a lower CPU clock frequency via dip switch would
help ?

Paul

 

[Daniel Mandic]

Hello Daniel,

Thanks for your messages!
Strange thing is, the CPU is tested on the card that I have, and it
works! But not on my P3B-F board.

I heard something about a reset of the P3B-F with a jumper, I must
try this. Another question is, must I change something on my
motherboard or bios to get it work?

And, I hear about a Slot-T converter... But I don't know if this
works...

Many questions and many options, and if one of these options give an
error of make a failure, the whole PC didn't startup... I can search
for another converter-card, maybe I have more luck with that...

Thanks,

Roland.

Hi Roland!



I always disable all at "Power Managment" in the BIOS. Every wake on
lan.. wake on this ...disable APM.. etc.
NT 5.1 shows then: ACPI Button is disabled. APM is not working properly
(redcrossed in devicemanager and a problem-device in sysinfo). You
cannot disable ACPI and APM totally!!! Maybe whith this Hardware Mod, I
have not checked yet, sorry.

If you would know what a pleasure it is, the system without them.
EX-ACPI IRQ9 is now the VGA (the fastest of IRQ9-11 - user IRQ´s). 9 is
already PC-AT. 8 to 0 is PC-XT.
You also have to study the BX Chipset.... There are plenty of pdf on
the internet.
How it places the IRQ´s. AFAIK, the BX is knitting everything out of
the ISA BUS, and not, other are just addressed and not bussed. It´s an
amazing piece of hardware, summing up 30 Years of IBM-PC Technology.
The faster parts (64bit AGTL+ and the 32PCI Bus) are not affected, in
terms of speed. Not even that, they are really fast.

Maybe you have to select your IRQ`s manually.... I don´t know.... It´s
also possible that XP needs ACPI (more IRQ`s) for MPS. NT4 should not,
as it don´t have ACPI at all. The fastest NT, IMO... you should give it
a try, for further extensive testing.



Best Regards,

Daniel Mandic

P.S.: The Maximum of cards is 2PCI, and 1ISA (maybe two, I have not
tried). And AGP, of course. The Chipset can do (compensate) more, sure,
but that´s without any Soft- or Hardware conflicts.

 

[Roland]

I go to try a Slot-T adapter, wich is made special for Tualatin-core based
CPU's.
Maybe will this work, if it work (or not) I place a message on this topic...

Roland.

 

[Stephan Grossklass]

I go to try a Slot-T adapter, wich is made special for Tualatin-core based
CPU's.

Good idea. If it still doesn't work with this adapter, change the power
supply, too.

Stephan

 

[Roland]

I try a Slot-T adapter, and... IT WORKS!

But, not in one time. I have reset the bios with the CLRTC-option on the
main-board.
This are two points on the mainboard, and with a small screwdriver I make
contact between this two points. After that, I changed my original Slot-1
CPU into the new one with converter. The PC starts normally and in the Bios
I must change the busmaster-frequency. But, my PC is running now at 1339 MHz
and works much faster than 750 MHz, hi.

Many thanks to everyone who helps me with this...

Roland.