A7N8X-E and Barton 3200+ runs at 1.75v - is this normal?

Discussion in 'Asus' started by DaveyB, Sep 4, 2004.

  1. DaveyB

    DaveyB Guest

    Hi there,

    I've installed an AMD 3200+ 400FSB Barton into an A7N8X-E DELUXE and all seems to be ok in day to
    day use but the BIOS has set the Vcore at 1.75v instead of 1.65v that the chip is supposed to run
    at. The Vcore BIOS setting won't go any lower than 1.75v for this CPU. Is this something to worry
    about?

    Cheers, Davy
     
    DaveyB, Sep 4, 2004
    #1
    1. Advertisements

  2. DaveyB

    Ed Guest


    Unless there's a bug with the E-Deluxe that doesn't sound right. 1.75v
    won't kill it but maybe the CPU is not a 3200+ but a 2500+ being made to
    run faster by using more vcore for stability. Barton 2500+ and 3200+ use
    the same 11x multiplier but run at different FSB MHz. (166 cs 200).

    If it's not a boxed retail CPU I guess it's possible the CPU could have
    been modified, did you take a good look at the CPU bridges?

    Ed
     
    Ed, Sep 4, 2004
    #2
    1. Advertisements

  3. DaveyB

    DaveyB Guest

    I did have a look at the cut bridges and they all look more or less the same.
    There's a guy over in alt.comp.hardware.overclocking.amd who took a photo of
    his /possible/ fake 3200+ and all the bridges are the same configuration as mine.

    Photo: http://members.cox.net/eharris2/CPUPic1.jpg

    I've tried different BIOS versions for the A7N8X but they all gave 1.75v as the lowest limit.

    Mine is an unboxed OEM chip off eBay ( shudder ). I guess that swings the odds against me!

    DaveyB
     
    DaveyB, Sep 4, 2004
    #3
  4. DaveyB

    Paul Guest

    This page has bridge info:

    http://fab51.com/cpu/barton/athlon-e23.html

    Erik's picture looks like label fraud, and maybe no bridges
    were modified. It would look like the FSB was incorrect, and
    most people simply adjust the clock in the BIOS to fix it.
    The multiplier on the 2500+ is the same as a 3200+, which means
    only the FSB needs to be modded. A remarker can either change
    the FSB, or just leave it, and the end user will assume the
    external clock always needs to be adjusted on a 3200+.

    This poster's Vcore is too high, just like yours.

    http://groups.google.com/groups?threadm=Xns9543EB0A86842tristanluscombehotma@194.117.143.38

    Look at both the top of the chip, for bridge mods, but also look at the
    bottom of the CPU, to see if there is any conductive paint on there. To
    see what pattern they might use to modify the CPU, use this page, and
    the Vcore menu. That will show you what pins to look for the conductive
    paint on the bottom of the CPU. As a limited number of connections would
    need to be modified, not all the mods pictured here would be needed.
    Some of them are already set by the bridges on top of the chip:

    http://www.ocinside.de/go_e.html?/html/workshop/pinmod/amd_pinmod.html

    If it is a hardware failure, it is a pretty strange one. I hope AMD
    isn't VID modding these themselves :)

    HTH,
    Paul
     
    Paul, Sep 4, 2004
    #4
  5. DaveyB

    Leadfoot Guest

    sounds like you were sold a modified chip

    Can you make the core voltage higher? Don't worry 1.85VDC won't hurt it

    What temp does it run at?

    Will it run Prime95 for 6-12 hours?

    If either of the latter two are a problem how hard will it be to get your
    money back?

    If they are not a problem then its probably not worth the hassle to take it
    back. It may not be the vendor who you purchased it froms fault. It could
    have been in 10 different hands between you and AMD
     
    Leadfoot, Sep 5, 2004
    #5
  6. DaveyB

    DaveyB Guest

    Thanks for those urls, very handy :)

    I've scanned the CPU and also used a 3D microscope to look at the bridges.

    It looks like third link up on L11 has been sliced.

    Full Pic: http://www.btinternet.com/~sector101/cpu/3200top.jpg
    Closeup: http://www.btinternet.com/~sector101/cpu/3200_L11.jpg

    Using the microscope thing I noticed some kind of paint joining two sets of pins on the underside
    of the cpu.

    Full Pic: http://www.btinternet.com/~sector101/cpu/3200bottom.jpg
    Closeup: http://www.btinternet.com/~sector101/cpu/3200bottommod.jpg

    The paint is well hidden under some tan colored coating, I didn't notice it until it was under the
    microscope.

    DaveyB
     
    DaveyB, Sep 6, 2004
    #6
  7. DaveyB

    Harry Guest


    You will notice the sqaure matrix on the right of the serial numbers.
    If you can, compare this to the one stamped on the CPU core. They
    should match. The sqaure is infact an AMD barcode

    Read here for more info

    http://www.xbitlabs.com/news/cpu/display/20031020183230.html

    cheers

    Harry
     
    Harry, Sep 7, 2004
    #7
  8. DaveyB

    Paul Guest

    Excellent! Thanks for educating us on the latest forgery methods!
    I guess there really is no substitute for just buying a 2500+
    and overclocking the FSB yourself. Maybe everyone who gets burned
    on this stuff, should email AMD and suggest they just sell
    2500+ from now on :)

    How is your negotiation going with the seller on Ebay ?

    Maybe it is up to the buyer to point out that they own a 3D
    microscope and know how to use it, before the product is sent :)

    Paul
     
    Paul, Sep 7, 2004
    #8
  9. DaveyB

    DaveyB Guest

    I've contacted the guy about getting a refund for the CPU and await his reply. He runs an online
    PC and PC parts shop and has sold bucketloads of 3200+ Bartons on eBay. I can't imagine that I'm
    the only one who got a modded 2500+ from him.

    We'll see how it turns out...

    DaveyB
     
    DaveyB, Sep 8, 2004
    #9
    1. Advertisements

Ask a Question

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

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.