GA 6VX7-4 F2

OK, this is a FAQ for the Asus P2B (440BX) family. There were problems at
one point
with multiplier values of 11, 12, and 14. Your multiplier is 1000/133 =
7.5,
so that shouldn't have been a problem.

http://homepage.hispeed.ch/rscheideg...grade_faq.html

The fact that you got 900MHz, should have stopped you right in your
tracks. If the multiplier on the processor really is 7.5, then 900/7.5 =
120.
If the input clock was 120, that is an "in-between" value for the clock.
It could be that 120 / 3 = 40Mhz for the PCI clock, which could have
caused your corruption on boot (some clock generators have a 120 / 3
setting and also a 120 / 4, so there are a couple possible outcomes).

The clock generator chip has a data sheet. Your user manual has a table,
but with four active switches at work, there are 16 possible frequency
settings, and your user manual doesn't document all of them.

If the clock generator is an ICS brand device, you can find an ICST.com
datasheet archived on here. I think mine might have been a 9150-08 or
something close to that. There are other brands of clockgen chips, but
these guys have a decent portion of market share. (Read the label on the
chip, to identify it - the user manual seldom mentions the identity of
what they used. Without an actual picture of your board, I cannot give
an exact location for it.)

http://web.archive.org/web/200412050....icst.com/pdf/

OK, I think I've got it. You user manual mentions "ICS 9248DF-39" in
the block diagram. This is the nearest datasheet. The chip is a 48 pin
device (24 legs per side):

http://web.archive.org/web/200410140...ics9248-39.pdf

Page 3 has a table. 1111 = 133Mhz. If you flipped all the switch
settings by accident, 0000 = 124MHz, with a PCI bus of 41.33MHz (ouch).
I think in fact, that the readout might have been closer to 930 MHz ?
Just a guess.

Double check your switch settings. I wouldn't let the POST finish and
boot into Windows, until you get the correct "1000" readout at the
beginning of POST. Turn off the power, and try the switches again,
before letting it get near the hard drives.

If you want a good way to test, download Knoppix (knopper.net) or
Ubuntu. The download is 700MB and is in ISO format. You use a CD
burning tool like Nero, and use the ISO to cause a bootable CD
to be burned. I have several different versions that I use
for booting potentially unstable systems. No software is installed
and no hard drive space is needed. In fact, you can disconnect
the hard drive if you want, while booting either of those OSes.
I do overclock testing with Knoppix, because I like watching the
text messages displayed, as stuff on the bus fails etc. You cannot
hurt a CD, like you can a hard drive, which is the nice feature of
booting a Linux LiveCD. (There is probably some minimal amount of
RAM needed to boot them - I have 1GB here, which is why I never
bothered to check the minimum.)

HTH,
Paul

Put the 600Mhz CPU back in and the pc ran fine no corruption of hard drive,
hanging or freezing.

Started again from afresh, took the 600 mhz cpu out, put the Ighz cpu in and
set the dip switches as follows: -
SW1 1 & 2 OFF, 3 ON & 4 OFF.
SW2 1 to 6 ALL OFF.

Checked memory sticks for 133Mhz as suggested by another poster, all 133mhz.

Disconnected hard drive (Scsi)

Switched on, booted up ok except still showing 900Mhz.

Tried to download Knoppix but page unavailable !?!

I shall be away from Thursday through to next Monday so plenty of time for
me to think this over.

Bruce.

Bruce Varney wrote:
"Invalid" wrote in message
...
In message , Bruce Varney
writes
I have checked and double checked the dip switches, I have even operated
the
switches quite a few times in case of dirty contacts, still report 900
Mhz.
I then tried setting S1 to 1&2 OFF 3 ON and 4 off and S2 all OFF, hasn't
made a blind bit of difference. I cannot believe that all the switches
are
defective in one way or another and since the only thing I have changed is
the processor chip I am starting to think that it is this that is
defective.
If I get time today I will put the 600 Mhz CPU back in and see what
happens.


Bruce, its always possible that the BIOS simply doesn't understand the CPU
speed settings, and the post is just wrong.

I (still) run a GA-6BXDS board from 1998/9. It's design spec is PII
processors 200-633Mhz. It is fitted with a later pair of P3/1GHz/100
processors. The processors are multiplier locked, and the frequency ratio
dip settings on the board have no effect whatsoever (as you appear to have
discovered!).

The post reports the processors as something like 958Mhz but all the
diagnostics show them running at 1Ghz (actually 998Mhz) . At the time I
seem to remember a discussion on here about the fact that the Bios used a
lookup table that only understood chips up to 958Mhz and it simply
reported the highest number in its list.

It may be that your Bios is in the same position.

Doesn't however solve your C Drive problem.

An alternative thought. What speed memory is fitted. Was the 600Mhz chip a
600/100 or a 600/133. If it was a 600/100 then you may have some 100Mhz
memory fitted. If the Bus speed is set to auto, then the 600/100 would run
just fine with 100Mhz memory, but the 1Ghz/133 would have all sorts of
problems.

Try running the board with the bus speed forced to 100Mhz - you will only
get 750Mhz out of the processor, but if that works, then I suspect you
have at least one stick of PC100 memory.

Hope this helps.

Regards
--
Peter R Cook

Hi Peter,

Thanks for your reply.

I checked the memory sticks, they are 128Mb 133mhz, the 600Mhz CPU is a
100Mhz FSB so I think your idea is a non starter, although it was worth
checking out though.

Bruce..


You have the 9248-39 clock generator datasheet and its table of
FS3-FS2-FS1-FS0 input settings. You should be able to see
similar entries in the motherboard manual for the SW1 and
in the 9248-39 clock generator table of values.

I'd try going back to 100/33 for the FSB and PCI. Then
see if it reads 750MHz or not. If it does, then you know
the multiplier is 7.5.

It could be, that it just isn't stable at 133 for some reason.
You should be able to run three sticks of RAM at PC133.

I'd start by booting up a memtest86+ boot floppy or CD (memtest.org),
and give that a shot, at both 750MHz (FSB100/PC100) and
then at 1000MHz (FSB133/PC133), and see which of those
test conditions are stable.

Paul

"Paul" wrote in message ...
Bruce Varney wrote:
"Invalid" wrote in message
...
In message , Bruce Varney
writes
I have checked and double checked the dip switches, I have even
operated the
switches quite a few times in case of dirty contacts, still report 900
Mhz.
I then tried setting S1 to 1&2 OFF 3 ON and 4 off and S2 all OFF,
hasn't
made a blind bit of difference. I cannot believe that all the
switches are
defective in one way or another and since the only thing I have changed
is
the processor chip I am starting to think that it is this that is
defective.
If I get time today I will put the 600 Mhz CPU back in and see what
happens.


Bruce, its always possible that the BIOS simply doesn't understand the
CPU speed settings, and the post is just wrong.

I (still) run a GA-6BXDS board from 1998/9. It's design spec is PII
processors 200-633Mhz. It is fitted with a later pair of P3/1GHz/100
processors. The processors are multiplier locked, and the frequency
ratio dip settings on the board have no effect whatsoever (as you appear
to have discovered!).

The post reports the processors as something like 958Mhz but all the
diagnostics show them running at 1Ghz (actually 998Mhz) . At the time I
seem to remember a discussion on here about the fact that the Bios used
a lookup table that only understood chips up to 958Mhz and it simply
reported the highest number in its list.

It may be that your Bios is in the same position.

Doesn't however solve your C Drive problem.

An alternative thought. What speed memory is fitted. Was the 600Mhz chip
a 600/100 or a 600/133. If it was a 600/100 then you may have some
100Mhz memory fitted. If the Bus speed is set to auto, then the 600/100
would run just fine with 100Mhz memory, but the 1Ghz/133 would have all
sorts of problems.

Try running the board with the bus speed forced to 100Mhz - you will
only get 750Mhz out of the processor, but if that works, then I suspect
you have at least one stick of PC100 memory.

Hope this helps.

Regards
--
Peter R Cook

Hi Peter,

Thanks for your reply.

I checked the memory sticks, they are 128Mb 133mhz, the 600Mhz CPU is a
100Mhz FSB so I think your idea is a non starter, although it was worth
checking out though.

Bruce..


You have the 9248-39 clock generator datasheet and its table of
FS3-FS2-FS1-FS0 input settings. You should be able to see
similar entries in the motherboard manual for the SW1 and
in the 9248-39 clock generator table of values.

I'd try going back to 100/33 for the FSB and PCI. Then
see if it reads 750MHz or not. If it does, then you know
the multiplier is 7.5.

It could be, that it just isn't stable at 133 for some reason.
You should be able to run three sticks of RAM at PC133.

I'd start by booting up a memtest86+ boot floppy or CD (memtest.org),
and give that a shot, at both 750MHz (FSB100/PC100) and
then at 1000MHz (FSB133/PC133), and see which of those
test conditions are stable.

Paul

I have run the PC with the CPU set for 750Mhz and ran the memtest program
for 24 hours with no errors occurring. It is now running at 1Ghz and
running the memtest program again for 24 hours.

What is the latest Bios version for the GA 6VX7-4 F2 Motherboard?

Bruce Varney wrote:

"Paul" wrote in message ...
You have the 9248-39 clock generator datasheet and its table of
FS3-FS2-FS1-FS0 input settings. You should be able to see
similar entries in the motherboard manual for the SW1 and
in the 9248-39 clock generator table of values.

I'd try going back to 100/33 for the FSB and PCI. Then
see if it reads 750MHz or not. If it does, then you know
the multiplier is 7.5.

It could be, that it just isn't stable at 133 for some reason.
You should be able to run three sticks of RAM at PC133.

I'd start by booting up a memtest86+ boot floppy or CD (memtest.org),
and give that a shot, at both 750MHz (FSB100/PC100) and
then at 1000MHz (FSB133/PC133), and see which of those
test conditions are stable.

Paul

I have run the PC with the CPU set for 750Mhz and ran the memtest program
for 24 hours with no errors occurring. It is now running at 1Ghz and
running the memtest program again for 24 hours.

What is the latest Bios version for the GA 6VX7-4 F2 Motherboard?

F4 is listed here, from 2003/10/07:

http://tw.giga-byte.com/Support/Moth...=GA-6VX7-4X-AP

Paul

"Paul" wrote in message ...
Bruce Varney wrote:

"Paul" wrote in message
...
You have the 9248-39 clock generator datasheet and its table of
FS3-FS2-FS1-FS0 input settings. You should be able to see
similar entries in the motherboard manual for the SW1 and
in the 9248-39 clock generator table of values.

I'd try going back to 100/33 for the FSB and PCI. Then
see if it reads 750MHz or not. If it does, then you know
the multiplier is 7.5.

It could be, that it just isn't stable at 133 for some reason.
You should be able to run three sticks of RAM at PC133.

I'd start by booting up a memtest86+ boot floppy or CD (memtest.org),
and give that a shot, at both 750MHz (FSB100/PC100) and
then at 1000MHz (FSB133/PC133), and see which of those
test conditions are stable.

Paul

I have run the PC with the CPU set for 750Mhz and ran the memtest program
for 24 hours with no errors occurring. It is now running at 1Ghz and
running the memtest program again for 24 hours.

What is the latest Bios version for the GA 6VX7-4 F2 Motherboard?

F4 is listed here, from 2003/10/07:

http://tw.giga-byte.com/Support/Moth...=GA-6VX7-4X-AP

Paul


F4 is for GA 6VX7-4X AP Motherboard, if I run this bios update I get BIOS ID
ERROR. I have also tried Bios F45 with the same results.