ST62K boots with incorrect fsb

Discussion in 'Shuttle' started by ozwald@operamail.com, Dec 28, 2004.

  1. Guest

    I have recently had an ST62K built for me by a firm in Plymouth, Devon
    UK. (But tha's another story.) I have a 3.0 GHz Prescott and I find
    that from time to time when starting up, the bios incorrectly sets the
    multiplier & FSB to 15 x 166 MHz instead of 15 x 200 MHz, giving me a
    2.49 GHz cpu clock. The RAM is always correctly recognised as 400 MHz
    dual channel.

    The error occurs when I explicitly set the multiplier & fsb to 15 x 200
    and also if I set it to "auto".

    I have flashed the bios to what I believe is the latest revision
    (suffix 'X'). (It had 'P' when I got it.)

    I have seen this error mentioned before; has anybody any idea if it is
    fixable, or will I just have to live with it?

    Also, what are good cpu temps? I am getting 54-56 on idle, and about 62
    on load. If I set the fan to 'ultra-low' in the bios, it hardly ever
    ramps up)
     
    , Dec 28, 2004
    #1
    1. Advertising

  2. Dodgy Guest

    On 28 Dec 2004 09:46:55 -0800, waffled on about
    something:

    >I have recently had an ST62K built for me by a firm in Plymouth, Devon
    >UK. (But tha's another story.) I have a 3.0 GHz Prescott and I find
    >that from time to time when starting up, the bios incorrectly sets the
    >multiplier & FSB to 15 x 166 MHz instead of 15 x 200 MHz, giving me a
    >2.49 GHz cpu clock. The RAM is always correctly recognised as 400 MHz
    >dual channel.
    >
    >The error occurs when I explicitly set the multiplier & fsb to 15 x 200
    >and also if I set it to "auto".
    >
    >I have flashed the bios to what I believe is the latest revision
    >(suffix 'X'). (It had 'P' when I got it.)
    >
    >I have seen this error mentioned before; has anybody any idea if it is
    >fixable, or will I just have to live with it?
    >
    >Also, what are good cpu temps? I am getting 54-56 on idle, and about 62
    >on load. If I set the fan to 'ultra-low' in the bios, it hardly ever
    >ramps up)


    I have a P4-2.8 (533FSB), sometimes it gets booted with 166Mhz clock
    instead of 133Mhz, amazingly enough it actually gets into windows and
    will work for a while before going a bit wobbly! (21x166=3.48Ghz!).

    Reported it to shuttle a few months ago, no reply, not even an
    automatic one... So no change there! I eventually wrote a little
    program (me being a programmer) that checked the speed and just
    rebooted the machine if it was wrong.

    Anyway, just found a new BIOS,

    FT62S00Z
    1. Fixed sometimes detect Fsb(front side bus) irregularly at boot up.

    http://global.shuttle.com/Download/Download_File.asp?Item=ST62K

    I'm gonna try it when I get home (well after I've found why the
    machine is now reporting NTLDR failure!)

    D0d6y.
    --
    MUSHROOMS ARE THE OPIATE OF THE MOOSES
     
    Dodgy, Jan 19, 2005
    #2
    1. Advertising

  3. Dodgy Guest

    On 24 Jan 2005 02:35:56 -0800, "Freddo_2005" <>
    wrote:

    >Here are your answers :
    >
    >1. CPU fsb problem : sorry, you have a defective motherboard... Search
    >sudhian forums for st62k
    >(http://forums.sudhian.com/messageview.aspx?catid=43&threadid=71495&highlight_key=y&keyword1=st62k)
    >
    >2. Temperatures : you could lower this by at least 5°C. Reboot, enter
    >the bios setup, advanced chipset config, and set the CPU voltage to
    >1.175V (instead of Auto). Then save and exit. Once logged into windows,
    >download prime95 and launch a 2hrs torture test. If ever you
    >exprerience a "fatal error", get back to the bios and increase CPU
    >voltage by one step. Don't worry: undervolting will not harm your CPU
    >in any way. Anyway, a 3.0GHz P4 Prescott should be perfectly stable at
    >1.175V (and even stable at 1.100V if you run it @2.5GHz). And DON'T
    >FORGET to disable the spectrum spreading (default 0.5%). You'll see
    >your shuttle dramatically cool down!!!


    Well I don't know about the OP, but the FT62S00Z bios update sorted my
    2.8Ghz p4... It hasn't booted with the wrong FSB since I flashed it.

    D0d6y.
     
    Dodgy, Jan 25, 2005
    #3
  4. Guest

    Here I am, the OP. I am not convinced the mobo is faulty. BIOS updates
    exist to correct mobo/cpu incompatibilities. I would not rush to call a
    mobo 'faulty' if it displayed a problem such as this. Hell, my old Abit
    and the Jetway before that were known to do things like this. I posted
    this same question on the Hexus.net SFF forum, and an ST62K user told
    me that the 00Z bios did not cure his case. (It did not cure mine
    either.)

    BUT!!! What did cure his, and also after I read it, mine, was a very
    bizarre thing. He wrote that switching the cpu fan connector from FAN1
    header to FAN3 stopped the wrong fsb problem completely. I don't know
    what made him try it. He said he did not know why it worked.

    So far I have had about 16 boots all correct.

    I had to change the settings in Speedfan of course.

    I have undervolted down to 1.200 and I am getting idle temp 39 C and
    100% cpu stabilises at 54 C after about 20 minutes slow climb, so I'm
    pleased considering it's a Prescott in a Shuttle case.
     
    , Jan 28, 2005
    #4
  5. Dodgy Guest

    On 28 Jan 2005 12:29:32 -0800, wrote:

    >Here I am, the OP. I am not convinced the mobo is faulty. BIOS updates
    >exist to correct mobo/cpu incompatibilities. I would not rush to call a
    >mobo 'faulty' if it displayed a problem such as this. Hell, my old Abit
    >and the Jetway before that were known to do things like this. I posted
    >this same question on the Hexus.net SFF forum, and an ST62K user told
    >me that the 00Z bios did not cure his case. (It did not cure mine
    >either.)
    >
    >BUT!!! What did cure his, and also after I read it, mine, was a very
    >bizarre thing. He wrote that switching the cpu fan connector from FAN1
    >header to FAN3 stopped the wrong fsb problem completely. I don't know
    >what made him try it. He said he did not know why it worked.
    >
    >So far I have had about 16 boots all correct.
    >
    >I had to change the settings in Speedfan of course.
    >
    >I have undervolted down to 1.200 and I am getting idle temp 39 C and
    >100% cpu stabilises at 54 C after about 20 minutes slow climb, so I'm
    >pleased considering it's a Prescott in a Shuttle case.


    Okay, you get the prize for weird!

    Glad you got it sorted, pity about shuttle's support, it really will
    effect my choice of compact PC next time.

    D0d6y.
     
    Dodgy, Jan 29, 2005
    #5
  6. Guest

    Dodgy wrote:
    > On 28 Jan 2005 12:29:32 -0800, wrote:
    >
    > >an ST62K user told
    > >me that the 00Z bios did not cure his case. (It did not cure mine
    > >either.)
    > >
    > >BUT!!! What did cure his, and also after I read it, mine, was a very
    > >bizarre thing. He wrote that switching the cpu fan connector from

    FAN1
    > >header to FAN3 stopped the wrong fsb problem completely. I don't

    know
    > >what made him try it. He said he did not know why it worked.
    > >
    > >So far I have had about 16 boots all correct.
    > >

    >
    > Okay, you get the prize for weird!
    >
    > Glad you got it sorted, pity about shuttle's support, it really will
    > effect my choice of compact PC next time.
    >
    > D0d6y.


    D0d6y, I spoke too soon! The next day the Zen was back to booting at
    2.49 GHz. I am wondering if the reason for the seeming cure was not so
    much the change of fan header, but the fact that the pc was moved, case
    was opened, the motherboard was maybe flexed slightly. So I am going to
    see if the RAM modules are seated properly.

    Still
     
    , Jan 29, 2005
    #6
  7. Dodgy Guest

    On 29 Jan 2005 01:59:03 -0800, wrote:

    >
    >Dodgy wrote:
    >> On 28 Jan 2005 12:29:32 -0800, wrote:
    >>
    >> >an ST62K user told
    >> >me that the 00Z bios did not cure his case. (It did not cure mine
    >> >either.)
    >> >
    >> >BUT!!! What did cure his, and also after I read it, mine, was a very
    >> >bizarre thing. He wrote that switching the cpu fan connector from

    >FAN1
    >> >header to FAN3 stopped the wrong fsb problem completely. I don't

    >know
    >> >what made him try it. He said he did not know why it worked.
    >> >
    >> >So far I have had about 16 boots all correct.
    >> >

    >>
    >> Okay, you get the prize for weird!
    >>
    >> Glad you got it sorted, pity about shuttle's support, it really will
    >> effect my choice of compact PC next time.
    >>
    >> D0d6y.

    >
    >D0d6y, I spoke too soon! The next day the Zen was back to booting at
    >2.49 GHz. I am wondering if the reason for the seeming cure was not so
    >much the change of fan header, but the fact that the pc was moved, case
    >was opened, the motherboard was maybe flexed slightly. So I am going to
    >see if the RAM modules are seated properly.
    >
    >Still


    If movement is the cause, and everything is seated, it's starting to
    sound like a dodgy motherboard again, maybe a hairline crack on a
    track, or a dry joint.

    D0d6y.
     
    Dodgy, Jan 30, 2005
    #7
  8. Guest

    > If movement is the cause, and everything is seated, it's sta­rting
    to sound
    > like a dodgy motherboard again, maybe a hairline crack­ on a
    > track, or a dry joint.


    > D0d6y.


    It looks like just changing the fan header is enough to cause a change
    temporarily.
    I found that changing the fan header back again from FAN3 to FAN1
    brought a number of correct boots but the next day (today) it is back
    to 15 x 166.
     
    , Jan 30, 2005
    #8
  9. Dodgy Guest

    On 30 Jan 2005 02:44:29 -0800, wrote:

    >> If movement is the cause, and everything is seated, it's sta­rting

    >to sound
    >> like a dodgy motherboard again, maybe a hairline crack­ on a
    >> track, or a dry joint.

    >
    >> D0d6y.

    >
    >It looks like just changing the fan header is enough to cause a change
    >temporarily.
    >I found that changing the fan header back again from FAN3 to FAN1
    >brought a number of correct boots but the next day (today) it is back
    >to 15 x 166.


    Me and my big mouth...

    Guess what I just noticed... Yeap, I've got the Z bios, and booted at
    166x21 (3.4ghz). I just happened to go into regedit for a nose, and
    saw the CPU speed. I don't get to see it very often as my monitor
    takes too long to wake up, so the PC is already past the front bios
    screen before I get a picture appear. I'm amazed my little 2.8Ghz P4
    runs so well at that speed!

    Back to the shuttle fault reporting... Oh joy.

    D0d6y.
     
    Dodgy, Jan 31, 2005
    #9
  10. Guest

    Update... I don't have to change the fan header or clear the CMOS (and
    go back to 00:00 Dec 21st 2004!!!); I spent half an hour experimenting,
    for example since i seemed to be stuck with a 166 MHz bus speed, I
    tried to see how far I could overclock in the BIOS, and it seemed
    stable at 10-12 MHz overclock, giving me about 2.8 GHz, but a couple of
    boots later I was was surprised to see my 200 MHz fsb back again, with
    the overclock it made Windows unstartable. After removing the overclock
    it stayed OK all day over about 6 reboots at intervals. Next day
    (today) I got the bad fsb at first boot, but after a couple of entries
    into the bios, changing things like RAM frequency up and down, I had a
    good fsb at boot which seems to have stuck again...

    I have twice made wildly optimistic overclock & RAM speed settings in
    the BIOS, rendering the system totally unbootable, so that I HAD to
    remove power and use the clear CMOS button, then load bios failsafe
    defaults, I both times got my 200 MHz fsb back.

    One can get awfully subjective about these things - prone to imagine
    that one's own efforts are changing the situation when really it's
    random or thermal. However I have a feeling (irrational?) that a
    straight series of boot cycles alone are not going to throw up a good
    fsb, I was trying that before & getting long runs of consistently
    erroneous starts.

    I wonder if the fix that Shuttle thought they had found for version Z
    needs to be made more aggressive or something.

    Or the voltages at boot are fluctuating too wildly for a correct
    hardware detection to happen every time. I could try taking loads off
    the psu like my 2nd hdd & see what happens...

    Incidentally I have undervolted down to 1.175 volts, and the rig seems
    stable and cool at 3 GHz - 39 C (idle) 54 C ( arduous mpeg4 encoding in
    VirtualDub) with Speedfan set for very quiet operation (desired temp
    62, min fan 1% max fan 15%)
    These temps are about 10 degrees cooler than setting cpu voltage to
    Auto in the BIOS.
     
    , Feb 1, 2005
    #10
  11. Dodgy Guest

    On 1 Feb 2005 04:07:34 -0800, waffled on about
    something:

    >Update... I don't have to change the fan header or clear the CMOS (and
    >go back to 00:00 Dec 21st 2004!!!); I spent half an hour experimenting,
    >for example since i seemed to be stuck with a 166 MHz bus speed, I
    >tried to see how far I could overclock in the BIOS, and it seemed
    >stable at 10-12 MHz overclock, giving me about 2.8 GHz, but a couple of
    >boots later I was was surprised to see my 200 MHz fsb back again, with
    >the overclock it made Windows unstartable. After removing the overclock
    >it stayed OK all day over about 6 reboots at intervals. Next day
    >(today) I got the bad fsb at first boot, but after a couple of entries
    >into the bios, changing things like RAM frequency up and down, I had a
    >good fsb at boot which seems to have stuck again...
    >
    >I have twice made wildly optimistic overclock & RAM speed settings in
    >the BIOS, rendering the system totally unbootable, so that I HAD to
    >remove power and use the clear CMOS button, then load bios failsafe
    >defaults, I both times got my 200 MHz fsb back.
    >
    >One can get awfully subjective about these things - prone to imagine
    >that one's own efforts are changing the situation when really it's
    >random or thermal. However I have a feeling (irrational?) that a
    >straight series of boot cycles alone are not going to throw up a good
    >fsb, I was trying that before & getting long runs of consistently
    >erroneous starts.
    >
    >I wonder if the fix that Shuttle thought they had found for version Z
    >needs to be made more aggressive or something.
    >
    >Or the voltages at boot are fluctuating too wildly for a correct
    >hardware detection to happen every time. I could try taking loads off
    >the psu like my 2nd hdd & see what happens...
    >
    >Incidentally I have undervolted down to 1.175 volts, and the rig seems
    >stable and cool at 3 GHz - 39 C (idle) 54 C ( arduous mpeg4 encoding in
    >VirtualDub) with Speedfan set for very quiet operation (desired temp
    >62, min fan 1% max fan 15%)
    >These temps are about 10 degrees cooler than setting cpu voltage to
    >Auto in the BIOS.


    I have yet to find a pattern... The thing that really annoys the hell
    out of me, is that even when you set the FSB to 133Mhz *manual* (which
    is what my little 2.8 wants), it still sometimes auto pick 166Mhz!

    Oh, and is there even a P4 that uses 166Mhz! Grrrrrr!

    D0d6y.
    --
    MUSHROOMS ARE THE OPIATE OF THE MOOSES
     
    Dodgy, Feb 1, 2005
    #11
  12. Dodgy <> wrote in message news:<>...
    > I have yet to find a pattern... The thing that really annoys the hell
    > out of me, is that even when you set the FSB to 133Mhz *manual* (which
    > is what my little 2.8 wants), it still sometimes auto pick 166Mhz!
    >
    > Oh, and is there even a P4 that uses 166Mhz! Grrrrrr!
    >
    > D0d6y.


    I have now found that watching the screen at boot time, and pressing
    the reset button (instantly) if the wrong cpu speed appears, seems to
    do the trick for me. The reset is very quick, the screen just jumps a
    bit that's all.

    There must be some kind of race conditions at the very moment of boot
    which cause this???
     
    Kapitan Klink, Feb 4, 2005
    #12
  13. Dodgy Guest

    On 3 Feb 2005 23:24:21 -0800, (Kapitan Klink)
    waffled on about something:

    >Dodgy <> wrote in message news:<>...
    >> I have yet to find a pattern... The thing that really annoys the hell
    >> out of me, is that even when you set the FSB to 133Mhz *manual* (which
    >> is what my little 2.8 wants), it still sometimes auto pick 166Mhz!
    >>
    >> Oh, and is there even a P4 that uses 166Mhz! Grrrrrr!
    >>
    >> D0d6y.

    >
    >I have now found that watching the screen at boot time, and pressing
    >the reset button (instantly) if the wrong cpu speed appears, seems to
    >do the trick for me. The reset is very quick, the screen just jumps a
    >bit that's all.
    >
    >There must be some kind of race conditions at the very moment of boot
    >which cause this???


    Unfortunately my monitor takes too long to wake up, so I miss that
    screen... All I can do is press pause at the right moment (at the
    wrong moment causes a keyboard error!), or, more often than not, check
    the registry once it's in windows and reboot pdq if it's wrong.

    HKEY_LOCAL_MACHINE\HARDWARE\DESCRIPTION\System\CentralProcessor\0\~MHz

    Very handy

    D0d6y.
    --
    MUSHROOMS ARE THE OPIATE OF THE MOOSES
     
    Dodgy, Feb 8, 2005
    #13
  14. > >I have now found that watching the screen at boot time, and pressing
    > >the reset button (instantly) if the wrong cpu speed appears, seems to
    > >do the trick for me. The reset is very quick, the screen just jumps a
    > >bit that's all.
    > >
    > >There must be some kind of race conditions at the very moment of boot
    > >which cause this???

    >
    > Unfortunately my monitor takes too long to wake up, so I miss that
    > screen... All I can do is press pause at the right moment (at the
    > wrong moment causes a keyboard error!), or, more often than not, check
    > the registry once it's in windows and reboot pdq if it's wrong.
    >
    > HKEY_LOCAL_MACHINE\HARDWARE\DESCRIPTION\System\CentralProcessor\0\~MHz
    >
    > Very handy
    >
    > D0d6y.


    That quick reset I mentioned must have been a fluke, or I
    misremembered it...

    I power up my monitor before starting the PC, but I have to then power
    up the PC within about 1 - 2 sec or the monitor says "no signal" &
    goes into standby & by the time it wakes up again the boot screen is
    past (but see the end of this post for my idea which makes the monitor
    unnecessary I think.)

    At boot the sequence of events is something like this:-
    (I have a thumb nail ready on the reset button.)

    1. Press power button.
    2. Fan runs at 100%.
    3. Memory test, EPA logo etc, cpu speed appear, fan quietens as BIOS
    takes control of speed.
    4. I see wrong cpu speed. (every time)
    5. Press reset button.
    6. Screen goes blank
    7. Steps 2 and 3 again.
    8. I see correct cpu speed. (every time)

    If your wrong boots are not happening every time you boot, you may
    prefer just checking in the registry or there is a good free app
    called cpu-z which does what you need (you could put it in your start
    up group).

    My problem is that just rebooting the computer from Windows doesn't
    seem to do the trick: I actually need to press the reset button. (I
    don't understand why this is.)

    Let's hope a BIOS update comes soon...

    With my situation (It always gets the wrong speed after a cold boot)
    if I had a monitor which was too slow to let me see the cpu speed in
    the BIOS screen, and given that a reset always puts things right, I
    would merely listen for the fan speed to quieten and press reset at
    that moment. Obviously pressing reset in Windows is to be deplored ;(
     
    Kapitan Klink, Feb 9, 2005
    #14
  15. Actually, I have refined the technique: I just press the reset button
    a couple of times while the BIOS is loading, during the period of high
    fan speed, it does not seem to affect the boot time, and I get the
    correct fsb every time.
     
    Kapitan Klink, Feb 10, 2005
    #15
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. W Luke

    ST62K CPU Monitoring

    W Luke, Apr 5, 2004, in forum: Shuttle
    Replies:
    2
    Views:
    316
    David Jones
    Apr 6, 2004
  2. mlj

    Zen (ST62K) and Windows 2003

    mlj, Apr 24, 2004, in forum: Shuttle
    Replies:
    2
    Views:
    300
  3. Robin Green

    ST62K : wierd FSB/CPU speed problem

    Robin Green, Oct 1, 2004, in forum: Shuttle
    Replies:
    0
    Views:
    349
    Robin Green
    Oct 1, 2004
  4. Kapitan Klink
    Replies:
    0
    Views:
    803
    Kapitan Klink
    Apr 24, 2005
  5. Replies:
    1
    Views:
    531
    John Doue
    Aug 20, 2007
Loading...

Share This Page