P2B Rev 1.02

  • Thread starter Thread starter Matt
  • Start date Start date
M

Matt

Hi List

I've been trying to fit a new cpu to
my aging p2b mobo, its a straight P2B rev 1.02
bios 1012 and the cpu is a P3-800/256/100/1.7V SL4BY

is there any compatibillity issue with such a combination ?
setting the jumper to 8x100 is unsucessfull, no Post at all

is there ather bios version than the 1012 that would fix the problem ?
is the 1.02 Rev to old for such a cpu ?
is there anybody out there ? ;p

Thanks
Matt~
 
Use Google to look at old discussions of this issue, but I *think* that
ver. 1.02 boards are incapable of supplying the required voltage.

-=-
Alan
 
some of the boards will work some will not. BTW is it a slot one or a socket
370 with and adapter. For the slot 1 it is dependent on the voltage
regulator chip. Socket 370 requires an adapter that has adjustable voltage
settings. If you have that set the voltage to 1.8
 
Alan Beagley said:
Use Google to look at old discussions of this issue, but I *think* that
ver. 1.02 boards are incapable of supplying the required voltage.

-=-
Alan

This page has some revision information. I think it says the board has
to be revision 1.12 or later, to give a voltage less than 1.8 volts.

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

To make this work, you have to change the VID code. I modified a slocket
to do this. In your case, you need to change the VID1 signal from a
logic 1 to a logic 0. So, grounding the VID1 signal would tell the
voltage regulator on the motherboard to supply 1.8 volts, instead of
the 1.7 volts the module is asking for. (VID codes are open collector
logic, with pullup resistors on the motherboard making the logic 1,
and the processor shorts the signal to ground to make a logic 0.
So, where there is a VID code of 1, the processor is basically
not connected to that VID signal.)

VID4 VID3 VID2 VID1 VID0 Vcore_voltage
0 0 1 1 1 1.70
0 0 1 0 1 1.80

To understand what is happening, here are the datasheets for an old
and new voltage regulator (this was used on my motherboard).

http://www.intersil.com/data/FN/FN4/FN4417/FN4417.pdf (HIP6004A)
http://www.intersil.com/data/FN/FN4/FN4567/FN4567.pdf (HIP6004B)

On the 6004A regulator, when asked for 1.7 volts, it outputs 0.0 volts
and so the computer won't post. The lowest valid voltage setting on the
6004A is 1.8 volts, so that is why the VID code must be changed to
accomodate the missing settings.

[ Changing a 1.7 or 1.75 volt Coppermine processor to 1.8 volts is
safe, but doing this to a Tualatin is not recommended, as a Tualatin
is a 1.5 volt processor with a maximum Vcore allowed of 1.75 volts. ]

So, now all you need to find, is instructions for the best way to
implement the modification :-)

HTH,
Paul
 
Matt said:
The CPU is a P3800/256/100/1.7 SL4BY SLOT1,
i've googled my way around and found differents result so far,
i tough the official ausus group was the place to go to straight
things out, AFAIK it may be the board (1.02) not going below
1.8V, since the Processor is a 1.7. hope i did not damage it
in an irreversible way.

P2B rev 1.02 definitely cannot go below 1.8v. The voltage regulator shut
down when your new CPU asked for 1.7v. There is no possibility the CPU
was damaged as a result.

It is quite easy and safe to run that CPU at 1.8v on your board - you
just need to put a small U-shaped wire in the Slot-1 connector to change
the voltage request to 1.8v. Details and pictures here:

http://tipperlinne.com/p2b-ds
Friend of mine have a P2B 1.10 wich i will try this weekend
ill keep you update on how it turns out !

Thanks
Matt

Use Google to look at old discussions of this issue, but I *think*
that ver. 1.02 boards are incapable of supplying the required voltage.

-=-
Alan


On 06/23/03 10:19 am Matt put fingers to keyboard and launched the
following message into cyberspace:


I've been trying to fit a new cpu to
my aging p2b mobo, its a straight P2B rev 1.02
bios 1012 and the cpu is a P3-800/256/100/1.7V SL4BY

is there any compatibillity issue with such a combination ?
setting the jumper to 8x100 is unsucessfull, no Post at all

is there ather bios version than the 1012 that would fix the problem ?
is the 1.02 Rev to old for such a cpu ?
is there anybody out there ? ;p

This page has some revision information. I think it says the board has
to be revision 1.12 or later, to give a voltage less than 1.8 volts.

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

To make this work, you have to change the VID code. I modified a slocket
to do this. In your case, you need to change the VID1 signal from a
logic 1 to a logic 0. So, grounding the VID1 signal would tell the
voltage regulator on the motherboard to supply 1.8 volts, instead of
the 1.7 volts the module is asking for. (VID codes are open collector
logic, with pullup resistors on the motherboard making the logic 1,
and the processor shorts the signal to ground to make a logic 0. So,
where there is a VID code of 1, the processor is basically
not connected to that VID signal.)

VID4 VID3 VID2 VID1 VID0 Vcore_voltage
0 0 1 1 1 1.70
0 0 1 0 1 1.80

To understand what is happening, here are the datasheets for an old
and new voltage regulator (this was used on my motherboard).

http://www.intersil.com/data/FN/FN4/FN4417/FN4417.pdf (HIP6004A)
http://www.intersil.com/data/FN/FN4/FN4567/FN4567.pdf (HIP6004B)

On the 6004A regulator, when asked for 1.7 volts, it outputs 0.0 volts
and so the computer won't post. The lowest valid voltage setting on the
6004A is 1.8 volts, so that is why the VID code must be changed to
accomodate the missing settings.

[ Changing a 1.7 or 1.75 volt Coppermine processor to 1.8 volts is
safe, but doing this to a Tualatin is not recommended, as a Tualatin
is a 1.5 volt processor with a maximum Vcore allowed of 1.75 volts. ]

So, now all you need to find, is instructions for the best way to
implement the modification :-)

HTH,
Paul
 
koop said:
If it is a Slot 1 PIII is *should* work, though in an over-voltage
condition, unless it is a Tualatin core though I don't think they went that
low in speed. I have a Slot 1 P2B rev 1.02 running an FCPGA Coppermine
PIII650@866 on an Iwill Slocket II, stilling running SETI night and day
since June 2000 - thought I would have found ET by now :-)

BigJIm had it right, the reason it works is because the FCPGA CPU uses the
voltage regulator on the slocket instead of the MB.

No it doesn't. There is no voltage regulator on the slotket. Setting the
jumpers to 1.8v simply tells the motherboard to supply 1.8v to the CPU.

The minimum voltage on
the slocket is still 1.8V so I am over spec there, but the presence of the
slocket voltage regulator satisfies the BIOS to continue booting.

No. The BIOS has nothing to do with it. When you set the jumpers to
1.8v, the CPU appears to be requesting a voltage the motherboard can
supply, so it does. If the CPU appears to be requesting a voltage lower
than the motherboard can supply, the voltage regulator shuts down. The
BIOS continues booting if it detects a CPU - which will only happen if
the voltage regulator is supplying power to the CPU.

IIRC I
 
Well thanks for clearing that up on the voltage regulator, you are entirely
correct. But obviously the BIOS does have something to do with it since
Matt's won't boot and as you say "The BIOS continues booting if it detects a
CPU - which will only happen if the voltage regulator is supplying power to
the CPU".

Koop
 
The P2B in any form (unless changing the onboard regulator) can't regulate down
to 1.8 volts or lower, even with a Slotkit--the system will not boot.
If you get a slotkit with a voltage regulator on it like the PowerLeap it will
work with the Tualatin cpu.
I am running a PowerLeap adaptor with a Tualatin 1.1.

Robert
 
Come on, guys. He has a slot 1 processor. It isn't going to boot in
this board unless he uses the wire in the slot idea above. This board
cannot provide 1.7 volts as it was built.
 
My older P2BLS had this issue; a Powerleap adapter supplied the needed
voltage and got me from P2-450 to P3-850
gr
 
The CPU is a P3800/256/100/1.7 SL4BY SLOT1,
i've googled my way around and found differents result so far,
i tough the official ausus group was the place to go to straight
things out, AFAIK it may be the board (1.02) not going below
1.8V, since the Processor is a 1.7. hope i did not damage it
in an irreversible way.

nah. that motherboard simply doesn't apply any voltage if it can't
supply the VID voltage the slot one wants.

I've got a p2b rev 1.02 too. I'm running a flip chip (FCPGA) P3 on a
simple slot one adapter card. No voltage regulator on the adapter
card, but it does have jumper pins to reassign the VID signal so as to
fool the motherboard about what voltage to supply to Intel's CPU.

The P2b rev 1.02 can run P3 coppermines (256 cache) by setting the cpu
voltage to 1.8 volts. I've been running for years now, it won't harm
anything.

You're problem is you've got a slot one P3 coppermine. There's no
jumpers for you to set the VID signal to 1.8 volts. You can use the
slot one cpu, but you would have to do a Mod to your current P3 VID
lines. It's not as hard as you might imagine if you know how to
solder, but if you're not a technician type, I wouldn't recommend it.

You can actually change the voltage regulator chip on the motherboard
to the newer rev level version and run a slot one, but it's a lot more
trouble than modifying your cpu VID lines. Again, it's for tech
hobbyists.

Bottom line, people with slot one cpu's and rev 1.02 usually sell the
slot one to someone else and pick up a flip chip P3. Today you can
use the tualatin P3 and celerons with the powerleap adapter cards. A
bit expensive for my taste. You can still find something like the P3
850 /256/100 flip chips out there relatively cheap.

I wouldn't toss the P2b, it's a very good motherboard, and I'd put in
a "cheap" P3 and run linux on it forever if nothing else.

I'm running windows2000 on mine with a 200 Gig WD hard drive off a
Promise controller card. I'm running a P3 overclocked at 133 FSB and
89mhz AGP buss. Just to give you some encouragement that the rev 1.02
can run very reliably with coppermines and 133 FSB. Note that you
need good pc133 memory and a video card that will handle the 89mhz AGP
bus speed if you want to run 133.

A good trick is to buy the 1Gig FCPGA P3 coppermine that runs at
133mhz FSB. If you have logistics problems running at 133, set your
jumpers for 100mhz operation. The 1Gig will run at 750mhz / 100mhz
FSB.

Another way to go is the 128k cache "coppermine" celerons at 1gig or
1.1 gig. These run 100mhz FSB. The problem is finding them these
days.

Whatever, update your bios to the the latest and greatest.

Good luck.


Friend of mine have a P2B 1.10 wich i will try this weekend
ill keep you update on how it turns out !

Thanks
Matt

Use Google to look at old discussions of this issue, but I *think* that
ver. 1.02 boards are incapable of supplying the required voltage.

-=-
Alan


On 06/23/03 10:19 am Matt put fingers to keyboard and launched the
following message into cyberspace:


I've been trying to fit a new cpu to
my aging p2b mobo, its a straight P2B rev 1.02
bios 1012 and the cpu is a P3-800/256/100/1.7V SL4BY

is there any compatibillity issue with such a combination ?
setting the jumper to 8x100 is unsucessfull, no Post at all

is there ather bios version than the 1012 that would fix the problem ?
is the 1.02 Rev to old for such a cpu ?
is there anybody out there ? ;p

This page has some revision information. I think it says the board has
to be revision 1.12 or later, to give a voltage less than 1.8 volts.

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

To make this work, you have to change the VID code. I modified a slocket
to do this. In your case, you need to change the VID1 signal from a
logic 1 to a logic 0. So, grounding the VID1 signal would tell the
voltage regulator on the motherboard to supply 1.8 volts, instead of
the 1.7 volts the module is asking for. (VID codes are open collector
logic, with pullup resistors on the motherboard making the logic 1,
and the processor shorts the signal to ground to make a logic 0.
So, where there is a VID code of 1, the processor is basically
not connected to that VID signal.)

VID4 VID3 VID2 VID1 VID0 Vcore_voltage
0 0 1 1 1 1.70
0 0 1 0 1 1.80

To understand what is happening, here are the datasheets for an old
and new voltage regulator (this was used on my motherboard).

http://www.intersil.com/data/FN/FN4/FN4417/FN4417.pdf (HIP6004A)
http://www.intersil.com/data/FN/FN4/FN4567/FN4567.pdf (HIP6004B)

On the 6004A regulator, when asked for 1.7 volts, it outputs 0.0 volts
and so the computer won't post. The lowest valid voltage setting on the
6004A is 1.8 volts, so that is why the VID code must be changed to
accomodate the missing settings.

[ Changing a 1.7 or 1.75 volt Coppermine processor to 1.8 volts is
safe, but doing this to a Tualatin is not recommended, as a Tualatin
is a 1.5 volt processor with a maximum Vcore allowed of 1.75 volts. ]

So, now all you need to find, is instructions for the best way to
implement the modification :-)

HTH,
Paul
 
nah. that motherboard simply doesn't apply any voltage if it can't
supply the VID voltage the slot one wants.

I've got a p2b rev 1.02 too. I'm running a flip chip (FCPGA) P3 on a
simple slot one adapter card. No voltage regulator on the adapter
card, but it does have jumper pins to reassign the VID signal so as to
fool the motherboard about what voltage to supply to Intel's CPU.

The P2b rev 1.02 can run P3 coppermines (256 cache) by setting the cpu
voltage to 1.8 volts. I've been running for years now, it won't harm
anything.

You're problem is you've got a slot one P3 coppermine. There's no
jumpers for you to set the VID signal to 1.8 volts. You can use the
slot one cpu, but you would have to do a Mod to your current P3 VID
lines. It's not as hard as you might imagine if you know how to
solder, but if you're not a technician type, I wouldn't recommend it.

It's a lot easier than you imagine - no soldering required:

http://tipperlinne.com/p2b-ds.htm
You can actually change the voltage regulator chip on the motherboard
to the newer rev level version and run a slot one, but it's a lot more
trouble than modifying your cpu VID lines. Again, it's for tech
hobbyists.

Bottom line, people with slot one cpu's and rev 1.02 usually sell the
slot one to someone else and pick up a flip chip P3. Today you can
use the tualatin P3 and celerons with the powerleap adapter cards. A
bit expensive for my taste. You can still find something like the P3
850 /256/100 flip chips out there relatively cheap.

I wouldn't toss the P2b, it's a very good motherboard, and I'd put in
a "cheap" P3 and run linux on it forever if nothing else.

I'm running windows2000 on mine with a 200 Gig WD hard drive off a
Promise controller card. I'm running a P3 overclocked at 133 FSB and
89mhz AGP buss. Just to give you some encouragement that the rev 1.02
can run very reliably with coppermines and 133 FSB. Note that you
need good pc133 memory and a video card that will handle the 89mhz AGP
bus speed if you want to run 133.

A good trick is to buy the 1Gig FCPGA P3 coppermine that runs at
133mhz FSB. If you have logistics problems running at 133, set your
jumpers for 100mhz operation. The 1Gig will run at 750mhz / 100mhz
FSB.

Another way to go is the 128k cache "coppermine" celerons at 1gig or
1.1 gig. These run 100mhz FSB. The problem is finding them these
days.

Whatever, update your bios to the the latest and greatest.

Good luck.



Friend of mine have a P2B 1.10 wich i will try this weekend
ill keep you update on how it turns out !

Thanks
Matt

Use Google to look at old discussions of this issue, but I *think* that
ver. 1.02 boards are incapable of supplying the required voltage.

-=-
Alan


On 06/23/03 10:19 am Matt put fingers to keyboard and launched the
following message into cyberspace:



I've been trying to fit a new cpu to
my aging p2b mobo, its a straight P2B rev 1.02
bios 1012 and the cpu is a P3-800/256/100/1.7V SL4BY

is there any compatibillity issue with such a combination ?
setting the jumper to 8x100 is unsucessfull, no Post at all

is there ather bios version than the 1012 that would fix the problem ?
is the 1.02 Rev to old for such a cpu ?
is there anybody out there ? ;p

This page has some revision information. I think it says the board has
to be revision 1.12 or later, to give a voltage less than 1.8 volts.

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

To make this work, you have to change the VID code. I modified a slocket
to do this. In your case, you need to change the VID1 signal from a
logic 1 to a logic 0. So, grounding the VID1 signal would tell the
voltage regulator on the motherboard to supply 1.8 volts, instead of
the 1.7 volts the module is asking for. (VID codes are open collector
logic, with pullup resistors on the motherboard making the logic 1,
and the processor shorts the signal to ground to make a logic 0.
So, where there is a VID code of 1, the processor is basically
not connected to that VID signal.)

VID4 VID3 VID2 VID1 VID0 Vcore_voltage
0 0 1 1 1 1.70
0 0 1 0 1 1.80

To understand what is happening, here are the datasheets for an old
and new voltage regulator (this was used on my motherboard).

http://www.intersil.com/data/FN/FN4/FN4417/FN4417.pdf (HIP6004A)
http://www.intersil.com/data/FN/FN4/FN4567/FN4567.pdf (HIP6004B)

On the 6004A regulator, when asked for 1.7 volts, it outputs 0.0 volts
and so the computer won't post. The lowest valid voltage setting on the
6004A is 1.8 volts, so that is why the VID code must be changed to
accomodate the missing settings.

[ Changing a 1.7 or 1.75 volt Coppermine processor to 1.8 volts is
safe, but doing this to a Tualatin is not recommended, as a Tualatin
is a 1.5 volt processor with a maximum Vcore allowed of 1.75 volts. ]

So, now all you need to find, is instructions for the best way to
implement the modification :-)

HTH,
Paul
 
Back
Top