Archived from groups: alt.comp.periphs.mainboard.asus (
More info?)
In article <nospam-2801050505500001@192.168.1.177>, nospam@needed.com
(Paul) wrote:
> In article <Brass.1jeajm@news.computerbanter.com>, Brass
> <Brass.1jeajm@news.computerbanter.com> wrote:
>
> > Recently purchased a P5GDC-V Deluxe. 3.2 Ghz, 1 GB DDR-2.
> > Trying toconfig 2 120 Gig Sata drives in a Raid 1 (Mirror)
> > configuration. .Install of Raid Driver/OS (XP Pro) goes off
> > without any issues. OnFirst boot, the system, hangs.
> > Specifically on mup.sys . Tireddisabling mup.sys as i'm not
> > planning to connect to a novell env, andit would proceed to
> > hang on the next file in line. Tried usingoriginal, older
> > raid drivers, updated the bios, swithed the SATAchannels,
> > multiple installs, disabled all onboard devices, no pci
> > cardinstalled, os installed on single sata drive without
> > raid successful.Any ideas would be appreciated. Solutions
> > Preferred.
> >
> > Thanks
> >
> > Brass-- Brass
>
> Dunno why, but this sounds a lot like a "WinXP SP2 microcode
> problem". This is caused by mixing SP2 with a BIOS that doesn't
> have a sufficient minimum revision of microcode for a
> Prescott processor.
>
> Find an OS you can get to boot, then follow this recipe:
>
>
http://groups.google.ca/groups?selm=%23D53KHviEHA.3712%40TK2MSFTNGP15.phx.gbl
>
> Latest beta BIOS:
>
http://www.asus.com.tw/support/download/selectftp.aspx?l1_id=1&l2_id=22&l3_id=3&m_id=1&f_name=1007-005.zip~zaqwedc
>
> Latest official release BIOS:
>
http://www.asus.com.tw/support/download/selectftp.aspx?l1_id=1&l2_id=22&l3_id=3&m_id=1&f_name=Gdcv1006.zip~zaqwedc
>
> Now, something interesting happened when I dissected 1007.005. I
> extracted the microcode segment from the BIOS, with AMIBCP75.exe .
> Normally, it would be a simple matter, to use "cmtc microcod.exe /store"
> to parse the microcode segment and spit out the microcode revisions.
> CTMC processed a few and then bailed.
>
> It seems either the microcode format has changed, to support newer
> processors, or the microcode segment is really screwed up. My bet
> is this is a new format. It looks like some microcode segments have
> a hex length of 0x0800 (2K) and others look longer.
>
> The format is documented here (PDF page 306):
>
http://www.intel.com/design/pentiumii/manuals/24319202.pdf
>
> To hand parse the file, I use a hex editor, and look for "0x01"
> on a 2KB (0x0800 hex) boundary. If I find one, chances are the
> other header fields will line up. This is what I find.
>
> 1007.005
>
> offset Family Update_revision
> 0x0000 0F25 2C
> 0x0800 0F37 02
> 0x1000 0F34 14
> 0x3000 0F34 08
> 0x4800 0F40 06
> 0x6000 0F41 06
> 0x7800 0F41 0D
> 0x9000 0F44 01
>
> 1006
>
> offset Family Update_revision
> 0x0000 0F25 2C
> 0x0800 0F31 0B
> 0x1800 0F32 07
> 0x2000 0F33 0B
> 0x3000 0F34 13
> 0x4800 0F34 08
> 0x6000 0F40 06
> 0x7800 0F41 09
> 0x8800 0F43 04
>
> Use the FrequencyID utility from Intel, and see what revision
> of processor and Update_revision claim to be running. You
> can actually look up the Family, by plugging the SSPEC off the
> box the processor came in, here. SSPEC is a five character
> field, like "SL123".
>
>
http://processorfinder.intel.com
>
> If the microcode revision turns out to be zero, that means
> that the BIOS didn't find a microcode to match your
> processor family.
>
> AFAIK, just running WinXP SP1 instead of SP2, will also fix this.
>
> HTH,
> Paul
There is a newer microcode spec on page 353 here. That
would explain the funny stuff I found above.
ftp://download.intel.com/design/Pentium4/manuals/25366814.pdf
Paul
Facebook
Twitter
Instagram