Does vcore affect overall stability or just CPU stability?

[Merc Zephyr]

Just wondering if an increase in vcore gets you stability, then was it definitely or most likely the CPU? Or is there a chance it could have been something else?

[OSUmaxx]

From my experience, if something is wrong besides the vcore, upping it will not help the problem. AFAIK it is only related to CPU stability.

[shiyan]

I'd agree. Most probably CPU.

[Deathknight]

Just cpu.

[Merc Zephyr]

Thank you for the replies!
Cheers

[dustybyrd]

i disagree with some of the previous posters...because upping the vcore on the cpu will ALSO require more power and thus may over-tax the power supply (particularly if it is a generic POS)...

so if the increased vcore increases the load on the psu too much rebooting may occur...

[shiyan]

I agree with what dustybyrd says, in that if your PSU isnt' strong enough, upping the vcore could actually decrease the system's stability.

But, from the way Merc Zephyr asked the question, it looks like he increased the vcore, after which his system is now stable.

Then he's asking whether the lack of stability before upping the vcore was just CPU related.

In which case I think it should be mainly CPU related.

[<ACIDBURN>]

so if you up the fsb should you also up the core volt ? how do you know when its to much voltage?

[c627627]

When you up the FSB and fail Prime95 Torture test, you up the Vcore gradually to see if you can pass the stability test at that same FSB speed but with higher voltage.

So if your default voltage is 1.60V for example, you'll find that you would get significatly higher stable FSB at 1.8volts instead of 1.6 volts.

The temperature rises of course.

You know it's too much voltage when you have to raise the voltage to get insignificant gains.

Also assuming your CPU is 'supposed' to last 10 years:


hitechjb:
A 10% increase in Vcore, would shorten the failure time to 83% of nominal failure time.
A 20% increase in Vcore, would shorten the failure time to 69% of nominal failure time.
A 30% increase in Vcore, would shorten the failure time to 59% of nominal failure time.
A 50% increase in Vcore, would shorten the failure time to 44% of nominal failure time.


So 30% is about 6 years (5.9) instead of 10... but you must keep temperatures in check or else your CPU is not going to last that long.

[<ACIDBURN>]

where can you get prim95 ???

[<ACIDBURN>]

I have a question i have a 1800 and it will go fine up to 2600 266 fsb but when i get to 333 fsb it wont start up no beep code no nothing just blah i have to reset the cmos just to get it back to working order any ideas ?

[L337 M33P]

www.mersenne.org/freesoft.htm

When you load it up, just select Test>torture test from the menu. If it bottoms out after less than 6 hours then you are either at the edge or need more Vcore.

[Deathknight]

Originally posted by dustybyrd
i disagree with some of the previous posters...because upping the vcore on the cpu will ALSO require more power and thus may over-tax the power supply (particularly if it is a generic POS)...

so if the increased vcore increases the load on the psu too much rebooting may occur...

Using that same logic adding a CD+RW also causes instability :D

There should be a general understanding that you need a PSU that can handle the load of your system. If you increase vcore and your stability decreases you either have created too much heat or as dustybyrd indicated your hitting your PSU too hard. I think the way the question was worded though he was looking to see if more vcore could have a positive impact on anything but the cpu stability. There are only 2 positive impacts that I know of. You get more stability for your CPU (if your cooling is up to the task) and you don't need to heat that room in the winter ;)

[<ACIDBURN>]

So you wold be freezing to the keyboard in winter ?

[<ACIDBURN>]

downlads not working ?

[Merc Zephyr]

But, from the way Merc Zephyr asked the question, it looks like he increased the vcore, after which his system is now stable.
Yep.
In which case I think it should be mainly CPU related.
Confirms what I have found with my testing.
There are only 2 positive impacts that I know of. You get more stability for your CPU (if your cooling is up to the task) and you don't need to heat that room in the winter
Could be a good thing during a Canadian winter.;) Could get some awesome cooling with the windows open.:D

Thanks again for replies. Cheers!

[yadda-yadda]

A 50% increase in Vcore, would shorten the failure time to 44% of nominal failure time.
Yeah right. Try to run Athlon XP at 2.9V and see how long it lasts. Try Intel at 2.9V and see if it boots.
What are these people smoking?

[DipStickTony]

the big IF is temperatures being in check. Running at 2.9v and 65c would kill the chip in like a week. Running it on a mach 1 or even very good watercooling is a totally different story.

[Merc Zephyr]

Running 2 instances of Prime95 with vcore set in bios to 1.675V, reading 1.58-1.61V under load and up to 1.71V idle, CPU temp is 44C. This is with a 2.60c@3.38ghz (not yet sure if stable) on a p4p800dlx. This temperature seems reasonable but I have to remember this board tends to give temps a little lower than many other boards.