NF70S does NOT undervolt at all!

Fever · Jul 12, 2003

I used to think it did, along with many other people. Readings from Winbond Hardware Doctor and MBM5 do not report the same VCore, so that's what made me decide to measure the actual vcore with a multimeter. I measured voltages ranging from 1.8v to 2.1v, and I was very surprized. What you set your voltage to in Softmenu III is EXACTLY what your chip is getting, so forget what every monitoring program tells you.

MBM 5.3.3.0 measures the VCore .09v to .11v too low, and Winbond Hardware Doctor measure the VCore .06v to .07v too low.

I hope I didn't damage my chip too much running it at 2.06v thinking it was really 1.95ish.

Just a little warning for you guys with a NF7-S Rev. 2.0.....

drewthomas14 · Jul 12, 2003

finally someone tested it. thanks

Fever · Jul 12, 2003

No problem.

james.miller · Jul 12, 2003

wow. so, i wonder what these people are really runinng when they claim [email protected]

drewthomas14 · Jul 12, 2003

james.miller said:
wow. so, i wonder what these people are really runinng when they claim [email protected]

what do you mean?

Fever · Jul 12, 2003

drewthomas14 said:
what do you mean?

He means when people say they're stable at 1.75v because that's what a program tells them, when in reality their VCore is higher then that.

Woody29 · Jul 12, 2003

james.miller said:
wow. so, i wonder what these people are really runinng when they claim [email protected]

i had often thought that, cos from what i've seen most are lucky to get 1.8v @ 2.4ghz.

pelikan · Jul 12, 2003

That's good info., thanks.
Did you find voltage fluctuation though? I figure that even if mbm5 reads wrong, the fluctuation it shows is probably there. In mbm5 my vcore fluctuates from 1.82-1.86.
Where exactly did you measure the vcore? I have read you can measure by the mosfets or on the copper coils.

Vertig0 · Jul 12, 2003

Do you think this applies to Vdimm readings as well? I mean, didn't everyone think that the NF7-S overvolted ram? Also, can you or someone else confirm this on another board or post instructions on how one would go about manually testing their Vcore?

Fever · Jul 13, 2003

About the VCore fluctuation, it does not at all. Maybe with some PSUs it might be different, but my Antec Trupower 480 didn't at all. The only "fluctuation" I saw was by .001v.

Also, Vcore doesn't change at all from idle to load.

To measure voltages yourself, check this thread out:
http://www.xtremesystems.org/forums/showthread.php?s=&threadid=14911

For the VCore reading I used the little, um, connections coming out of the mosfets. There's 3 of them on each mosfet. I didn't try reading any of the other voltages, but I suspect that they are like the VCore; exactly the same as they're set in your BIOS.

If you don't know how to use a multimeter, all you have to do is ground the negative wire (black) by making it touch some piece of metal (i used my case) and touch the measuring point with the positive wire (red). Be VERY careful that you don't slip and touch something else on your board, or else you might kill it. I made that mistakes and sparks went a flying. Thankfully my board didn't even complain.

IR1 · Jul 13, 2003

what does the actual bios report the vcore being? Undevolted or what it should be.

I noticed the undervolt effect in MBM5 and inside the bios. Also... when I put the computer on load the vcore drops .04. Normally it reports .03 lower, so on load its a whole .07 lower than what I set in bios.

I have a Sparkle 400watt. These suckers are known for their good stability, all my other rails are rock solid. The motherboard must just be reporting the vcore incorrectly if what your saying is true.

I never had this undervolt effect with my a7v8x even on load.

Woody29 · Jul 13, 2003

well, i hope we'll get to the bottom of this.
we need more people to post with the psu (wattage)brand and their vcore stable or not, to get some comparison.

well i'll be picking up my new CWT 550ADP tomorrow i'm hoping this will fix the vcore fluctuation problem and give me more stability at higher vcore settings

pelikan · Jul 13, 2003

Fever- thanks for the link. One more question, if you don't mind: You said you have no vcore fluctuation when you measure it directly. Does mbm5 show fluctuation?
It would be good to know the degree of inaccuracy in mbm5; whether its just voltage level being off, or innacurate fluctuation as well.
I'll test my board and post results.

Fever · Jul 13, 2003

IR1 said:
what does the actual bios report the vcore being? Undevolted or what it should be.

I noticed the undervolt effect in MBM5 and inside the bios. Also... when I put the computer on load the vcore drops .04. Normally it reports .03 lower, so on load its a whole .07 lower than what I set in bios.

I have a Sparkle 400watt. These suckers are known for their good stability, all my other rails are rock solid. The motherboard must just be reporting the vcore incorrectly if what your saying is true.

I never had this undervolt effect with my a7v8x even on load.

The bios is off by about .03, so even that's not perfect. It's better than MBM5 or Winbond, though. Yeah, Sparkles are very good PSUs.

Woody29 said:
well, i hope we'll get to the bottom of this.
we need more people to post with the psu (wattage)brand and their vcore stable or not, to get some comparison.

well i'll be picking up my new CWT 550ADP tomorrow i'm hoping this will fix the vcore fluctuation problem and give me more stability at higher vcore settings

What kind of PSU do you have now? If you're talking about VCore fluctuations with software monitors, it won't fix it. I get fluctuations also with my Antec TP 480W, but not when measured with the multimeter.

pelikan said:
Fever- thanks for the link. One more question, if you don't mind: You said you have no vcore fluctuation when you measure it directly. Does mbm5 show fluctuation?
It would be good to know the degree of inaccuracy in mbm5; whether its just voltage level being off, or innacurate fluctuation as well.
I'll test my board and post results.

No problem.

Don't worry about having more questions, that's why we're here.

Yes, both MBM and Winbond show fluctuations. MBM usually fluctuates by about .05 while Winbond only fluctuates by .01, but still shows the wrong voltage.

Fever · Jul 13, 2003

Whoops, ignore this.

Woody29 · Jul 13, 2003

hi fever....

at the moment i'm using a fotron 400w psu the problem is the higher i set the vcore the more it fluctuate, especially towards 1.8 and higher. when i set to say 1.825 in the bios hardware dr. reads 1.76 when under stress like prime sometimes lower.
but how much does your vcore drop or fluctuate by? cos your using a decent psu does it drop as much say like mine does.
so for me to actually get 1.8 vcore i really have to go to 1.85 or higher but then the siren kicks in when i go above 1.825 vcore

Fever · Jul 13, 2003

Woody29 said:
hi fever....

at the moment i'm using a fotron 400w psu the problem is the higher i set the vcore the more it fluctuate, especially towards 1.8 and higher. when i set to say 1.825 in the bios hardware dr. reads 1.76 when under stress like prime sometimes lower.
but how much does your vcore drop or fluctuate by? cos your using a decent psu does it drop as much say like mine does.
so for me to actually get 1.8 vcore i really have to go to 1.85 or higher but then the siren kicks in when i go above 1.825 vcore

That's exactly how mine is. You have to ignore what Hardware Dr. says, since it's wrong. When you set 1.85 in BIOS, you're really getting 1.85v. Do you have a multimeter to test your vcore with?

Woody29 · Jul 13, 2003

yeah i have one, but do not know where to poke it though

Fever · Jul 13, 2003

Take a look at this thread, woody.

http://www.xtremesystems.org/forums/showthread.php?s=&threadid=14911

Woody29 · Jul 13, 2003

thanks fever, i'll check it out now

NF70S does NOT undervolt at all!

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Member

Similar threads