• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Oops! We've Set XP Mobile vCore Wrong All This Time!

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Quentin

Member
Joined
Jul 12, 2003
Location
NorthWest USA
It appears we're wrong assuming vCore settings in the BIOS for desktop CPUs translate properly for XP Mobiles! In fact we need to use this simplified table below.

IE: to set a mobile XP to 1.6 volts the BIOS must be set to 1.65. If you set the BIOS to 1.6 the mobile actually would have a vCore of 1.5!

Look what happens if you set the BIOS to 1.85 - the mobile is at 2.0v! And if you set the BIOS to 1.525 the mobile is at 1.35. Forget running it at 1.45 in the BIOS, the mobile would be starved for volts at 1.275.

No wonder so many people have reported XP mobiles run real cool at low voltages but when you crank up the vCore they get RED HOT! (And all these volt mods to get more vCore were unnecessary if you have a mobile.)

Anybody have more info on this?

BIOS vCore. / . Actual
Setting for . / . vCore of the
DesktopXP . / . XP-Mobile CPU

1.850 <=======> 2.000
1.825 <=======> 1.950
1.800 <=======> 1.900
1.775 <=======> 1.850
1.750 <=======> 1.800
1.725 <=======> 1.750
1.700 <=======> 1.700
1.675 <=======> 1.650
1.650 <=======> 1.600
1.625 <=======> 1.550
1.600 <=======> 1.500
1.575 <=======> 1.450
1.550 <=======> 1.400
1.525 <=======> 1.350
1.500 <=======> 1.300
1.475 <=======> - N/A
1.450 <=======> 1.275
1.425 <=======> 1.250
1.400 <=======> 1.225
1.375 <=======> 1.200
1.350 <=======> 1.175
1.325 <=======> 1.150
1.300 <=======> 1.125
1.275 <=======> 1.100
1.250 <=======> 1.075
1.225 <=======> 1.050
1.200 <=======> 1.025
1.175 <=======> 1.000
1.150 <=======> 0.975
1.125 <=======> 0.950
1.100 <=======> 0.925

This comes from a thread over at aoaforums:
http://www.aoaforums.com/forum/showthread.php?t=30322

And the full VID/VCC_CORE table can be seen at:
http://fab51.com/cpu/barton/athlon-e20.html
 
Last edited:
if this is in fact true i'm shocked that it's only now coming out. i would have thought some of the more learned ppl here would have pointed this out. i'm assuming also that no one on xtreme sys posted about this either cause plenty of ppl here frequent there as well.
 
i also think if this was true im sure someone at XS would of picked this up along time ago

altho it could be true but it doesnt affect me as i dont have a XP-M
 
Well, so far only "Mad Dogs and Englishmen" have come out (sorry, just too good to resist!) :D

Anybody else have something to say - especially those with a little EE experience?
 
i don't think you need any ee experience, just real life experience. i have used bartons (desktop and mobile) for a long time. even though the measurements of vcore you get in windows are inaccurate, the relative values can be used. I ran a desktop barton and got the exact same readings with the exact same boards with the exact same bios vcore settings as with the mobile barton. I have never heard of this theory before, and I laughed when I read it. The vcore has been measured with the motherboard's sensors and with multimeters and I've never heard anything like this.
 
If this were the case, my mobile running at 1.95v on water would be like 2.3v actual. I think it might be dead then....
 
hmm interesting...maybe asking amd to confirm this would be good? Becauase as moz_21 said, many people run there mobiles at 1.95v-2.1v and they are still running fine.
nice find though.
 
moz_21 said:
If this were the case, my mobile running at 1.95v on water would be like 2.3v actual. I think it might be dead then....

Yeah if we were to extrapolate from the table - pretty darn hot:


2.000 <=======> 2.300
1.975 <=======> 2.250
1.950 <=======> 2.200
1.925 <=======> 2.150
1.900 <=======> 2.100
1.875 <=======> 2.050

Don't know if that's valid, though but it follows the pattern in the upper half of the table
 
well i am still using a T-Bred B and a Spitfire so dont fee like your bringing up the rear
 
Ever heard of Motherboard Undervolting? My Abit Nf7-Sv2.0 will always supply .25 Under the vCore I set in the BIOS even If Im using a desktop chip.
 
mudpark41 said:
Ever heard of Motherboard Undervolting? My Abit Nf7-Sv2.0 will always supply .25 Under the vCore I set in the BIOS even If Im using a desktop chip.
My NF7-S undervolts my mobile a LOT; almost .1v!:(
 
So let me get this right. Anything under 1.7, the motherboard undervolts a mobile chip. And anything over 1.7 it overvolts. I wonder how this translates to the infamous NF7-S v2.0 boards, which are notorious for undervolting. Mine undervolts at about 2.5v as well. Does this mean that 1.95v is the breaking point for my board? I am supprised that with all of the multimeters in this forum, that these results have not come out before.
 
Ok so this must mean that XP mobiles are the all time heaviest duty CPU becuase people have put 2.1v through them on water for a weak and still are under 50c which if I believe your chart would mean that they are pushing close to 2.5v through there CPUs.

Also when Fab51 did that article they never told you which motherboard they are using. Or how they calculated that table. Also I don't notice major temp jumps when running my CPU @ over 1.7v However at over 1.9v it goes up pretty fast but not untolerably.

Finally whats this at the top of the Fab51 article supposed to mine.
The following report is applicable the Barton and Thoroughbred.
However, my presumed part is contained. Please confirm for yourself.

This report is inapplicable to the Locked Athlon. Please refer to the Locked Athlon.

Then its possible this may also be old information as if you look at the bottom of the article you see
Copyright 2002 ita. All rights reserved. Never reproduce or republicate without written permission.

However hope fully OCdetective can tell us for sure as he seems to be able to find out anything AMD.
 
:bang head

now if we could just figure out the right voltages for ram .. :rolleyes:

got any tables to show us ?

ah forget it .. i'll use a multimeter.. and please dont ream me about the accuracy of a multimeter i know its a risk.. lol
 
Back