GA-P35-DQ6 & QX6800

On Thu, 21 Jun 2007 10:35:37 -0400, Paul <nospam.TakeThisOut@needed.com> wrote:

>Robert E. Wilson wrote:
>> Hello,
>>
>> Has anyone tried using an Intel QX6800 processor with the GA-P35-DQ6
>> motherboard?
>>
>> I just installed it with an ATI HD2900XT graphics card and it won't
>> POST. No beeps either. The case fans and graphics card fan spin. The
>> CPU fan (Thermaltake V1) gives just a small jerk (as it trying to
>> start) the instant the power is turned on, but doens't spin at
>> all--until I turn the power off, at wich time it lights up and spins
>> for about half a second. If connected to 12V through a molex adapter,
>> it spins fine. I've checked installation of the fan, graphics card,
>> etc. I've triple checked all power connections (Ultra 1000W)
>>
>> I'll try swapping the graphics card tonight with the one from my
>> wofe's computer (my only other PCI-E card). I'll even try another
>> power supply. I don't have any other DDR2 RAM at present, other than
>> the stick currently on the MB (more is coming "soon").
>>
>> Any ideas?
>>
>> Thanks,
>> Bob
>
>Try it without video and RAM and see if the computer speaker beeps.

Thank for the reply. Have done this--no beeps.



>
>The manual shows:
>
>ATX 2x4 12V processor power
>24 pin main power connector
>A 1x4 Molex connector next to the main power
>
>Are all of those currently connected ?

Tried with all connected (and without the 1x4 Molex. No difference.


>
>I'd get out the multimeter and check the voltages on those
>three connectors. You should be able to probe the metal on
>the pins while they are all plugged in.
>
>I guess this is your Ultra 1000W here ?
>
>http://www.jonnyguru.com/review_details.php?id=103
>http://www.ultraproducts.com/product_resources.php?cPath=52&pPath=632&productID=632

Yes, that's the PS.



>
>With only one video card present, the 1x4 Molex shouldn't need
>to be connected. The 24 pin has two 6 amp pins, which should be
>enough to keep a single 4 amp video card happy, plus the fans.
>So you could test without the 1x4 Molex, as a separate test case.

Done that.


>
>If you run out of other things to try, pull the motherboard
>from the case, and sit it on top of an insulating surface,
>like a telephone book with cardboard cover. That will eliminate
>shorting to the bottom of the board, as a potential problem.
>
> Paul

Will be getting to that point soon!
 >> Stay informed about: GA-P35-DQ6 & QX6800 

On Thu, 21 Jun 2007 10:35:37 -0400, Paul <nospam.TakeThisOut@needed.com> wrote:

>
>I'd get out the multimeter and check the voltages on those
>three connectors. You should be able to probe the metal on
>the pins while they are all plugged in.
>

All voltages on the back of the power connectors are good, except
there is no -5V (not even a wire in the connnector from the power
supply. I did try another PS that has the -5V line, but no difference
(-5V present, but the system wouldn't start, or even beep).

In addition (since my other reply), I've tried another video card,
checked that the processor and heatsink are properly installed, and
run with the motherboard out of the case (on a piece of plywood) to
rule out shorts in the case. No change with any of these attempts.

I've not been able to try another stick of RAM, but I doubt that's the
problem as I'm not even getting any error beeps.

It seems I'm left with either a bad MB, a bad CPU, or a CPU that isn't
compatible with my MB. If the latter, I hop Gigabyte updates the BIOS
SOON to support it! Though, I'd still have to get a compatible CPU to
use to flash the BIOS!

Thanks for you suggestions. Any further ideas appreciated!

--Bob
 >> Stay informed about: GA-P35-DQ6 & QX6800 

Robert E. Wilson wrote:
> On Thu, 21 Jun 2007 10:35:37 -0400, Paul <nospam.DeleteThis@needed.com> wrote:
>
>> I'd get out the multimeter and check the voltages on those
>> three connectors. You should be able to probe the metal on
>> the pins while they are all plugged in.
>>
>
> All voltages on the back of the power connectors are good, except
> there is no -5V (not even a wire in the connnector from the power
> supply. I did try another PS that has the -5V line, but no difference
> (-5V present, but the system wouldn't start, or even beep).
>
> In addition (since my other reply), I've tried another video card,
> checked that the processor and heatsink are properly installed, and
> run with the motherboard out of the case (on a piece of plywood) to
> rule out shorts in the case. No change with any of these attempts.
>
> I've not been able to try another stick of RAM, but I doubt that's the
> problem as I'm not even getting any error beeps.
>
> It seems I'm left with either a bad MB, a bad CPU, or a CPU that isn't
> compatible with my MB. If the latter, I hop Gigabyte updates the BIOS
> SOON to support it! Though, I'd still have to get a compatible CPU to
> use to flash the BIOS!
>
> Thanks for you suggestions. Any further ideas appreciated!
>
> --Bob

The CPU support list doesn't have the QX6800.
http://www.gigabyte-usa.com/Support/Motherboard/CPUSupport_List.aspx?P...uctID=2

Looking on processorfinder.intel.com, the QX6800 is similar to a QX6700,
only 2.93GHz. The only difference, appears to be the default multiplier.
It is the same stepping as the QX6700.

The -5V signal has been removed from power supplies for a while, and
a new motherboard should not be using it. If they need a negative voltage,
as bias for something, they can use the =12V.

Depending on your budget, I can see two things I might purchase.

1) Celeron 331 is supported. Newegg has one for $35. Use for motherboard
testing or BIOS flashing. The BIOS chips on your board look to be soldered
to the motherboard, but the pictures I'm looking at are not very high
resolution.
2) A PCI POST card. Newegg has a cheap looking one from Startech for $36.75
These are also available on Ebay, and can be quite cheap there (they
probably ding you for shipping). Some cards have voltage monitor LEDs
(likely simple window comparators per rail). Some cards have two edge
connectors, but how many ISA equipped motherboards do people debug now ?

The only benefit of the PCI POST card (placed in PCI slot #1), is for
activity detection. If the "FF" or "00" code stays on the display when
you start the motherboard, then no BIOS code is being executed. If the
code changes, it means some amount of BIOS code has executed. In the
newsgroups at least, the POST card doesn't seem to have much diagnostic
value, as the codes on the display are "progress" codes and not
"error" codes. The codes tell you what routine just executed, but in
the past, looking up the codes in available tables, generally doesn't
point at some defective piece of hardware. So in that sense, the POST
code card is no better than a single LED that says "I executed some
code".

I'd have made something of the fan symptoms, but the fan could be
connected to some kind of speed control, and that could be what
has given the weird fan speed at startup. The fan symptoms are not
necessarily indicating that the rail powering the fan is defective.

Paul
 >> Stay informed about: GA-P35-DQ6 & QX6800 

Thanks.

I've queried Gagibyte about the QX6800 support, but who knows how
quickly they'll respond. I expect that the board does or will support
the processer as it has the new P35 chipset and is advertised as
"optimized" for quad-core processors. I've also found some later BIOS
versions (probably beta) for this board. Of course, at the present, I
can't flash it!

I did order another processor (one on the supported list), but doubt
it'll arrive before next week.

I don't have the POST card--haven't seen the need due to the
limitations you mentioned. I figure that if nothing change when I try
a different processor, then the problem lies in the MB. If it works,
then it's either a bad QX6800 or the QX6800 isn't supported (yet).

I think the fan issue is that it is not being "turned on" since it is
in a speed control socket, and it spins just fine if plugged into a
non-speed controlled socket. Probably the boot up hasn't proceeded
far enough to turn it on. I'm guessing here, but since it almost
starts to spin when the system is first turned on, and then actually
does spin for a very short while as it is turned off, I'm wondering if
the fan header defaults to full voltage, but that the MB throttles it
down if it isn't detecting higher temps at the CPU. If so, it would
initially shut the fan down as the CPU starts cold (the brief attempt
to spin would be due the the brief application of power before the
circuit throttles it). Since the CPU appears to not be doing
anything, it never heats up to activate the fan.

BTW, the BIOS is soldered.

Thanks again for your suggestions.

--Bob

On Fri, 22 Jun 2007 00:51:13 -0400, Paul <nospam.RemoveThis@needed.com> wrote:

>Robert E. Wilson wrote:
>
>The CPU support list doesn't have the QX6800.
>http://www.gigabyte-usa.com/Support/Motherboard/CPUSupport_List.aspx?ProductID=2505
>
>Looking on processorfinder.intel.com, the QX6800 is similar to a QX6700,
>only 2.93GHz. The only difference, appears to be the default multiplier.
>It is the same stepping as the QX6700.
>
>The -5V signal has been removed from power supplies for a while, and
>a new motherboard should not be using it. If they need a negative voltage,
>as bias for something, they can use the =12V.
>
>Depending on your budget, I can see two things I might purchase.
>
>1) Celeron 331 is supported. Newegg has one for $35. Use for motherboard
> testing or BIOS flashing. The BIOS chips on your board look to be soldered
> to the motherboard, but the pictures I'm looking at are not very high
> resolution.
>2) A PCI POST card. Newegg has a cheap looking one from Startech for $36.75
> These are also available on Ebay, and can be quite cheap there (they
> probably ding you for shipping). Some cards have voltage monitor LEDs
> (likely simple window comparators per rail). Some cards have two edge
> connectors, but how many ISA equipped motherboards do people debug now ?
>
>The only benefit of the PCI POST card (placed in PCI slot #1), is for
>activity detection. If the "FF" or "00" code stays on the display when
>you start the motherboard, then no BIOS code is being executed. If the
>code changes, it means some amount of BIOS code has executed. In the
>newsgroups at least, the POST card doesn't seem to have much diagnostic
>value, as the codes on the display are "progress" codes and not
>"error" codes. The codes tell you what routine just executed, but in
>the past, looking up the codes in available tables, generally doesn't
>point at some defective piece of hardware. So in that sense, the POST
>code card is no better than a single LED that says "I executed some
>code".
>
>I'd have made something of the fan symptoms, but the fan could be
>connected to some kind of speed control, and that could be what
>has given the weird fan speed at startup. The fan symptoms are not
>necessarily indicating that the rail powering the fan is defective.
>
> Paul
 >> Stay informed about: GA-P35-DQ6 & QX6800 

Just an update for anyone interested, or those searching with a
similar problem in the future.

I purchased an Intel E4300 CPU, which is on Gigabyte's supported
processor list for this MB. It also wouldn't work--same response as
the QX6800. I therefore concluded that the MB was defective and had
it exchanged.

I rec'd the new MB today (incidently, there was an additional piece of
foam packaging supporting the northbridge heatsink that was not
present in my original one--and my original heatsink was bent some!

I also rec'd a reply from Gigabyte regarding support of the QX6800.

Anyway, the E4300 worked just fine in the new MB. I used it to flash
the BIOS to version F5g (sent to me by Gigabyte in order to support
thye QX6800-which I believe is a beta bios). The QX6800 is now
working in the MB.

In summary, the original board was defective. The Intel QX6800 is
supported, but requires a bios upgrade to at least F5g.

Thanks for all input!

Bob


On Thu, 21 Jun 2007 08:51:22 -0400, Robert E. Wilson
<not.me.DeleteThis@adelphia.net> wrote:

>Hello,
>
>Has anyone tried using an Intel QX6800 processor with the GA-P35-DQ6
>motherboard?
>
>I just installed it with an ATI HD2900XT graphics card and it won't
>POST. No beeps either. The case fans and graphics card fan spin. The
>CPU fan (Thermaltake V1) gives just a small jerk (as it trying to
>start) the instant the power is turned on, but doens't spin at
>all--until I turn the power off, at wich time it lights up and spins
>for about half a second. If connected to 12V through a molex adapter,
>it spins fine. I've checked installation of the fan, graphics card,
>etc. I've triple checked all power connections (Ultra 1000W)
>
>I'll try swapping the graphics card tonight with the one from my
>wofe's computer (my only other PCI-E card). I'll even try another
>power supply. I don't have any other DDR2 RAM at present, other than
>the stick currently on the MB (more is coming "soon").
>
>Any ideas?
>
>Thanks,
>Bob
 >> Stay informed about: GA-P35-DQ6 & QX6800