Usual limits of AGP/PCI bus overclock?

Archived from groups: alt.comp.hardware.overclocking.amd,alt.comp.hardware.overclocking (More info?)

Unfortunately my VIA K8T800 MB chipset doesn't allow for AGP/PCI bus
locking.

I can run my system at 76AGP and 38PCI (229fsb) but it'll crash/lockup
after a 30 sec or so. 220 is stable all day.

Is there a general rule of thumb for the max or near unstable limits of
oc'ing the AGP/PCI bus or a way to determine(besides pulling components)
if its a hardware piece unable to handle the AGP/PCI oc and not the true
cpu limit?
7 answers Last reply
More about usual limits overclock
  1. Archived from groups: alt.comp.hardware.overclocking (More info?)

    On Sat, 17 Apr 2004 08:07:26 -0500, Destroy <no@thanks.com> wrote:

    >Unfortunately my VIA K8T800 MB chipset doesn't allow for AGP/PCI bus
    >locking.
    >
    >I can run my system at 76AGP and 38PCI (229fsb) but it'll crash/lockup
    >after a 30 sec or so. 220 is stable all day.
    >
    >Is there a general rule of thumb for the max or near unstable limits of
    >oc'ing the AGP/PCI bus or a way to determine(besides pulling components)
    >if its a hardware piece unable to handle the AGP/PCI oc and not the true
    >cpu limit?

    Due to the weird ratio system they use, there should be some magic
    numbers at which the AGP/PCI speeds will be normal. Try around 233,
    234 Mhz. You can drop your cpu multiplier from 10 to 9 if it's too
    fast for the cpu.
  2. Archived from groups: alt.comp.hardware.overclocking.amd,alt.comp.hardware.overclocking (More info?)

    "Destroy" <no@thanks.com> wrote in message
    news:yX9gc.59440$YC5.13042@twister.rdc-kc.rr.com...
    > Unfortunately my VIA K8T800 MB chipset doesn't allow for AGP/PCI bus
    > locking.
    >
    > I can run my system at 76AGP and 38PCI (229fsb) but it'll crash/lockup
    > after a 30 sec or so. 220 is stable all day.
    >
    > Is there a general rule of thumb for the max or near unstable limits of
    > oc'ing the AGP/PCI bus or a way to determine(besides pulling components)
    > if its a hardware piece unable to handle the AGP/PCI oc and not the true
    > cpu limit?
    >

    As a general rule of thumb I would say about 10-12% increase on the fsb is
    all the pci/agp can take.

    Depending on what cpu it is - (assuming a t'bred 'b' core or a barton) the
    cpu should get up to around 2400mhz - providing everything such as, vcore,
    memory speed, memory voltage, memory timing, etc is set right - so if you
    are short of that figure, and everything else is fine then you know it is
    the pci/agp. If you can change the multiplier on the cpu, or temporarily
    use an unlocked one, set the multiplier stupidly low and see how far the fsb
    can go - that way you know it is not a cpu limit.
  3. Archived from groups: alt.comp.hardware.overclocking.amd,alt.comp.hardware.overclocking (More info?)

    Destroy wrote:

    > Unfortunately my VIA K8T800 MB chipset doesn't allow for AGP/PCI bus
    > locking.
    >
    > I can run my system at 76AGP and 38PCI (229fsb) but it'll crash/lockup
    > after a 30 sec or so. 220 is stable all day.
    >
    > Is there a general rule of thumb for the max or near unstable limits of
    > oc'ing the AGP/PCI bus or a way to determine(besides pulling components)
    > if its a hardware piece unable to handle the AGP/PCI oc and not the true
    > cpu limit?

    You can yank out everything but the bare-essentials, but then you'd still
    don't know if it's the CPU, southbridge, IDE, RAID, etc. To my knowledge,
    there is no way to know if it's a CPU limit, except by putting the CPU in a
    mainboard which does support PCI/AGP locking.

    BTW, what does the memory bus do when you increase the host-clock, seeing as
    how the memory-controller is located inside the CPU?

    --
    To send me, Halfgaar, email, remove remove from my email address.
  4. Archived from groups: alt.comp.hardware.overclocking.amd,alt.comp.hardware.overclocking (More info?)

    "Destroy" <no@thanks.com> wrote in message
    news:yX9gc.59440$YC5.13042@twister.rdc-kc.rr.com...
    > Unfortunately my VIA K8T800 MB chipset doesn't allow for AGP/PCI bus
    > locking.
    >
    > I can run my system at 76AGP and 38PCI (229fsb) but it'll crash/lockup
    > after a 30 sec or so. 220 is stable all day.
    >
    > Is there a general rule of thumb for the max or near unstable limits of
    > oc'ing the AGP/PCI bus or a way to determine(besides pulling components)
    > if its a hardware piece unable to handle the AGP/PCI oc and not the true
    > cpu limit?

    It really depends on the components you have. I have known some Radeon 9700
    graphics cards not to even tolerate 68MHz!

    BTW, I don't know if you know but the new "K8T800 Pro" chipset has a PCI and
    AGP lock. Should be interesting!

    Chip
  5. Archived from groups: alt.comp.hardware.overclocking.amd,alt.comp.hardware.overclocking (More info?)

    Hippy Paul wrote:


    > Depending on what cpu it is - (assuming a t'bred 'b' core or a barton) the
    > cpu should get up to around 2400mhz - providing everything such as, vcore,
    > memory speed, memory voltage, memory timing, etc is set right - so if you
    > are short of that figure, and everything else is fine then you know it is
    > the pci/agp. If you can change the multiplier on the cpu, or temporarily
    > use an unlocked one, set the multiplier stupidly low and see how far the
    > fsb can go - that way you know it is not a cpu limit.

    The K8T800 is an Athlon 64 chipset.... So no T-Bred or Barton. And one also
    shouldn't speak of an FSB, but a hypertransport bus.

    --
    To send me, Halfgaar, email, remove remove from my email address.
  6. Archived from groups: alt.comp.hardware.overclocking.amd,alt.comp.hardware.overclocking (More info?)

    Destroy wrote:
    > Unfortunately my VIA K8T800 MB chipset doesn't allow for AGP/PCI bus
    > locking.
    >
    > I can run my system at 76AGP and 38PCI (229fsb) but it'll crash/lockup
    > after a 30 sec or so. 220 is stable all day.
    >
    > Is there a general rule of thumb for the max or near unstable limits of
    > oc'ing the AGP/PCI bus

    Not really... it's not really the bus that fails at those speeds, but the
    devices attached to it. So it could be any device attached to your PCI or
    AGP bus, which is just about everything!

    > or a way to determine(besides pulling components)
    > if its a hardware piece unable to handle the AGP/PCI oc and not the true
    > cpu limit?

    Only if you have control of the multiplier, and even then it's not a sure
    sign. By lowering the multiplier you will lower the overall speed of the
    CPU, but will be able to leave the FSB alone... if it works in this
    situation, it's likely the CPU, if not, it's likely the components attached
    to the PCI (or AGP) bus.

    Ben
    --
    A7N8X FAQ: www.ben.pope.name/a7n8x_faq.html
    Questions by email will likely be ignored, please use the newsgroups.
    I'm not just a number. To many, I'm known as a String...
  7. Archived from groups: alt.comp.hardware.overclocking.amd (More info?)

    "Halfgaar" <voor_achter@yahoo.remove.com> wrote in message
    news:c5s8jt$c7s$1@news4.tilbu1.nb.home.nl...
    > Hippy Paul wrote:
    >
    >
    > > Depending on what cpu it is - (assuming a t'bred 'b' core or a barton)
    the
    > > cpu should get up to around 2400mhz - providing everything such as,
    vcore,
    > > memory speed, memory voltage, memory timing, etc is set right - so if
    you
    > > are short of that figure, and everything else is fine then you know it
    is
    > > the pci/agp. If you can change the multiplier on the cpu, or
    temporarily
    > > use an unlocked one, set the multiplier stupidly low and see how far the
    > > fsb can go - that way you know it is not a cpu limit.
    >
    > The K8T800 is an Athlon 64 chipset.... So no T-Bred or Barton. And one
    also
    > shouldn't speak of an FSB, but a hypertransport bus.
    >
    > --
    > To send me, Halfgaar, email, remove remove from my email address.


    oops - thanks for pointing that out
Ask a new question

Read More

AMD Overclocking Hardware PCI Bus