Archived from groups: alt.comp.periphs.mainboard.asus (
More info?)
"Paul" <nospam@needed.com> wrote in message
news:nospam-1110040705500001@192.168.1.177...
> In article <MPG.1bd412ec80718d719896a9@news.easynews.com>, Peregrin
> <yogi_berra@neverspamnever.org> wrote:
>
> > Hi all,
> >
> > I'm about to turn on my new machine, which will be my first try with a
> > Mobile Athlon in a Desktop MOBO. Is there a chance I may fry the CPU
> > before I can get in the bios to adjust the Vcore? The Vcore in the BIOS
is
> > set initially to AUTO. What is that going to be, with a mobile CPU? And
> > will the CPU be recognized correctly? If not, what is likely to happen?
> >
> > Is there anything I should/can do before I turn on the box, or just go
> > ahead and boot, and cross my toes while I use my fingers to get in the
> > bios and set the Vcore to say 1.5V, as fast as I can?
> >
> > And can I rely on the onboard indicated CPU voltage and temp., until
I'll
> > have installed WinXP and then MBMON?
> >
> > Thanks very much for any and all advice
> >
> >
> Mine started at 1.575. I wouldn't worry about it too much. I've
> given my 2600XP-M more voltage, but it doesn't seem to like too
> much.
>
> Start here - "Code to CORE Voltage Definition"
>
>
http://fab51.com/cpu/barton/athlon-e23.html
>
> I think my part number had "FQQ" as part of the string of
> characters, and the voltage is coded as "1.575/1.450(Q)".
> (The HSF is still on it right now, so I cannot verify the
> model number for you.) What that means is, the Mobile VID
> pins are set to 1.45V if plugged into a Mobile ready
> motherboard, but when plugged into the A7N8X-E, the voltage
> shows up as 1.575 nominal. The actual Vcore circuit always
> overvolts by a tiny amount, so the live reading in the Power
> Monitor will probably be 1.6V or so.
>
> In terms of voltage, there is VCC_CORE_Nom and VCC_Core_max.
> In this case VCC_CORE_Nom is 1.45 for the Q part, and
> VCC_Core_Max is 0.05V more. On top of that, you are allowed
> an extra 0.5V, to get to Absolute_Max. So, for the model 10
> Barton datasheet I have on the screen here, your maximum overvolt
> would be 1.45V+0.05V+0.50V = 2.0V. The fact that the actual
> motherboard gives the processor 1.6V when it starts up, is
> a non-issue. When I say 2.0V, that is the voltage at which the
> processor is damaged, so don't actually go there. I find that
> before you even get close to 2.0V, you'll probably find the
> extra voltage is counterproductive. So far, I think 1.65 was
> pretty good, and 1.7 was OK, and you might not need more than
> that (it just makes it run hotter). My typical delta_T from
> CPU to case is about 13C-14C, using a Zalman 7000A AlCu running
> at full speed. The motherboard draws 15A from +5V when benching
> and 10A from +5V when idle (under overclocked conditions). The
> idle current is cut in half when running at 800MHz, and idling
> at 800MHz, the delta_T can be as little as 3C.
>
> I think the temps are realistic, as nothing on the motherboard
> gets that hot.
>
> My biggest problem so far has been RAM, and I've discovered my
> cheap RAM is Memtest86 good but Prime95 bad. Assuming your RAM
> is good, I think you'll enjoy this combo. Run Prime95 over night
> to be absolutely sure everything is working right - Memtest86
> is not enough, as I've just learned. Memtest86 is still valuable
> for stuck bits, but I think Prime95 may be slightly better at
> finding other problems.
>
> HTH,
> Paul
I wouldn't be running any AMD at 2.0 volts without onions!