• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

cannot alter VCORE on evga...

S

Skychan

n00b
Joined
Dec 25, 2005
Messages
23
Ok guys, this issue might need some fairly expert help! I do not come empty handed though, since my last thread "Worth the tradeoff?" I finally understood the last thing I needed to understand about OCing and I have read OCing guides, etc, etc. In fact I have read threads regarding the OCing of systems almost identical to mine.

Yet, I cannot achieve even a slight OC.

I underclock everything:
LDT:600 (HTT Freq)
Mem FSB: 166
Command Rate: 2T
Mem timings: 3 clocks (instead of 2.5)
which all works fine and reports correctly in CPU-Z

then I increase the FSB (HTT actually) from 200 to 210mhz and it boots fine, but dies in Prime95 very quickly (but only in one copy... i think usually Affinity 0).

Now the vcore is highly unstable, indicated in CPU-Z as jumping around from 1.324 to 1.39, but the default should be 1.350.

Here is the kicker, and the reason for my thread title:
as soon as I alter the Vcore to anything (even setting it manually to 1.350) my machine won't even fuxing post! I figured increasing the vcore might help stabilize it for prime95, but no it seems i can't. Maybe an RMA is in order. Im hoping you experts out there can point out that im just doing something dumb.

My system:
4400+
evga SLI mATX 131-K8-NF44-AX mobo (i've even read a HUGE thread on this mobo here at the [H]ard Forum)
2*1024 OCZ platinum DDR400
Enermax Liberty 500w
QPack

P.S. I quote from Operaman who has this mobo and Opty 165 and Mushys:
"CPU FSB: 278 Mem FSB: 166 Voltage @ 1.525v (though it is seriously undervolting) and mem voltage @ 2.85v (Mushy is rated to 2.9 so what the heck)"
I never messed with my mem voltage....

http://www.hardforum.com/showthread.php?t=1023993&page=7&pp=20&highlight=sli
 
That's what he's got... He said his LDT was 600.

Hows your PSU?

I think you need to *find* something damaged with your board and RMA it ^_^
 
ya, i've got a great PSU, Enermax Liberty 500w.

So you really think this is something wrong with my mobo?

It would be nice if someone out there with this exact mobo could tell me whether they have experienced or not experienced this problem of setting the Vcore. That way I can decide if I need to RMA it.

(side note: what can i "say" is wrong with the mobo to rma it?)
 
Are you sure its the board? Try swaping out the processor first (pick one up from CompUSA and return it when your done testing) to make sure you don't just have a dud CPU.

If you still can't overclock even 10MHz, call up EVGA and tell them whats going on. There's no way their Quality Control's margin of error is as small as 10MHz, so you'll probobly get your RMA .
 
yeah, what are you going to say though? I mean, it runs fine at stock so thats all they should care about right? Guess its time to make up some stuff :)
 
jcll2002 said:
yeah, what are you going to say though? I mean, it runs fine at stock so thats all they should care about right? Guess its time to make up some stuff :)
Click to expand...
A) It’s an EVGA product, they should be expecting people to OC them.
B) The BIOS settings go to, what, like 300MHz?
C) You can make a stink about only having 10MHz of headroom between a stable system and information loss.
 
well first off I can't just grab another CPU like that as i live in canada and im guessing the nearest CompUSA is kinda far...

in any case for now what i'll do is put everything back to stock and run Prime 95 for like 8 hrs on both cores. I ran it nearly 2 hours on stock and didnt run into any errors. But if after 8 hrs i get an error at stock settings.... well something needs to be RMAed.

The question is what? In this thread we've discussed the mobo and CPU as being possibilities. It may also be possible that this board is being finicky and doesnt like my OCZ ram. Any thoughts?

Also, which of the torture test settings in P95 is best for this testing? I've been doing the top one, the "Small fft..." or whatever it says.

thx, i'll revive this thread late tonight after my long stress test.
 
thx for the advice, Uknown One. Hopefully that will work, although I'll be really pissed if i have to RMA this sucker!

Oh and that reminds me! According to CPU-Z the Vcore is swimming all over the place between 1.324 and 1.376. As I understand it the default Vcore settings is supposed to be 1.350. Is it normal for the Vcore to swim around so much? If not, then since my PSU is high quality, it must be a faulty mobo, correct?
 
My vcore jumps from 1.38 to 1.42 all the time. That's pretty normal :). The Large FFT setting will Stress the RAM. Failing at 8hrs prime, on stock settings is pretty weak... I mean, TECHNICALLY it should never fail. Lots of my stock systems have failed once in awhile, but then went on to priming for DAYS... Try it again, it might be a fluke.

Does that board have a particular NB fan that... well... might suddenly stop working?

**EDIT** not that I condone such an activity, especially against a great company like eVGA. Try calling them and explaining the situation first before anything suddenly fries it's self...
 
Voltage fluctuations are far too common. My rig is anywhere from 1.5 to 1.52 for CPU, and I'm sure the ram voltages do the same thing. I REALLY doubt that a fluctuating voltage would cause instability with such a low overclock, and not any instability at stock. Technically speaking it's possible that your CPU is just that close to the limit at stock, but I'd doubt it. My guess is if a CPU was that close, they would bump it down to a lower speed and package it as a slower processor, rather than possibly have to deal with an RMA for instability later on.

What happens if you lower all the settings, without raising the FSB at all?
 
Thx for the input you guys!

First off, at stock settings it never fails Prime95, ran it for like 24hrs.

Guess what, I solved the problem of not being able to POST when setting the Vcore to anything simply by changing the cpu multiplier from Default to 11X. When set to 11X I can increase the Vcore and POST and boot windows and get stable at FSB of 210 (which is still a very low OC)

Two problems remain:

1) Regardless of how high I set the Vcore in the bios (I've tried everything from 1.35 to 1.55) the mobo will NOT go higher than 1.40 as indicated by nVidia Monitor and CPU-Z.

2) At 220 FSB with vcore 1.40 the Affinity 0 core always always fails Prime 95 about 5 seconds into the test. Affinity 1 was fine for 5 minutes and Im guessing it would and should be stable for hours even at FSB of much higher than 220. CPU Affinity 0 appears to be defective.

So do I have a defective CPU, or have I got a bad HSF seating? I could try putting on my Blue Orb now and see if that makes a difference.
 
I'm getting quite interested in your problem, it seems very odd to me. I looked up an article on the PSU as I wasn't very familiar with it, seems to indicate that any voltage fluctuations are not the fault of the PSU. link. That, and the fact that that other guy you quoted in your first post was complaining of undervolting seems to me to imply that this board's voltage regulation SUCKS. I think the CPU is fine, but the board is bad. Do you have any other motherboards around to test the CPU/mem with?
 
Arcygenical said:
Failing at 8hrs prime, on stock settings is pretty weak... I mean, TECHNICALLY it should never fail. Lots of my stock systems have failed once in awhile, but then went on to priming for DAYS...
Click to expand...
My system reboots itself sometimes at stock settings, at 2.85ghz it's rock stable though...

What can I say, it's a DFI board.. it HAS to be overclocked.
 
Hey Factor, thx for your concern.

Unfortunately this is my first new age computer (PCI-E, S939) so I have no other motherboards.

I was aware of the risk of getting this "difficult" board, I had read that entire 14 or so page-long thread concerning this motherboard, but several ppl succeeded in OCing and i really wanted SLI.

This mobo isn't exactly "undervolting", but more like that it just plain won't bother changing its Vcore setting past 1.40 regardless of what you set it to (and the settings stays at what u set). Nvidia Monitor indicates that the mobo is set to 1.40 and both the bios and CPU-Z report fluctuations near 1.40. And the good news is that now its a MUCH more stable voltage! Fluctuates between 1.392 and 1.408 (once in a while as low as 1.376).

I am personally leaning towards the theory that one of the cores is busted up. I am hoping that better cooling is really all i need. I'll try the Blue Orb hopefully tomorrow.

Load (both cores) temps at 2.3ghz were 48c.
 
Just a note, don't rely on software ever, it doesn't even really give you a rough estimate of what voltage you have.

Buy a multimeter if you are concerned about voltage.
 
Thx for the comment Borgschulze, you may very well be correct.

At this point its been established that this mobo cannot increase the voltage anymore, but the X2s are low voltage CPUs. Its not exactly a great mobo for OCing.

I'm determined to succeed though :)
 
Tried my Blue Orb II CPU HSF.

Didn't really improve temperatures, and again CPU-0 dies instantly in Prime95 at FSB of 220mhz.

I haven't had any advice from anyone (though I would love some) but I think this means the CPU is defective and I should RMA it. What should I say though? I mean, it runs absolutely perfectly fine at stock settings.

:confused:
 
You must log in or register to reply here.
Back
Top