I meant 1.15v, not 1.5v. Oops

[Dankk]

Got my new i7-4790k rig the other day. One of the first things I did was manually set the CPU VCORE in the BIOS; I intended to set it at 1.15v. Instead, I accidentally set it to 1.5v. Oops.

While it was under this voltage, I installed Windows, ran all of the updates, and installed a few games (and played a couple of them). It wasn't until I ran an OCCT stress test when I noticed that something was wrong. It automatically stopped only seconds into the test because the core temperature was going over 85 C (thank goodness for that).

So I checked the voltage. Noticed that I had set it to 1.5. I fixed it, and now it's running great.

I guess I just have some anxiety over knowing that I had my CPU overvolted that high for any period of time. What are the chances that I did any damage to my CPU during those few hours when I was installing stuff?

Does the i7-4790k have any safety nets that make it throttle past a certain point? I'm worried it was getting past 80C, perhaps past 90C while installing Windows. (But then again I'm not sure. And I wasn't running any stress-tests after all, up until that last point.)

I'm happy to report that, with a *proper* voltage and a good closed-loop water cooler, it barely reaches 70 C during stress testing. So that's nice.

 

[hunkeelin]

Don't worry. You cpu weren't in fullload during ur software installation. Even if you put 1.55 it won't damage your cpu that much. Let's assume what you did, did infact reduce your cpu life span. Let's say from 12 years to 10 years... do you honestly think you will keep your cpu over 6+ ? Don't worry bro.

 

[the_stigma]

I wouldn't worry too much about it. Unless you reach a critical threshold where it literally kills the CPU as you are watching even extreme voltage needs some time to do anything significant. While 1.5 is definitely too high for Haswell for actual use, in all likelyhood it is still a ways of "instakill" range, and I'd be very surprised if just a few hours of non-synthetic varied load probably had caused any measurable degradation.

This is ultimately one of those "no one can tell you for sure" questions, but it seems unlikely enough that it's not worth worrying about IMO.

-Stigma

 

[Yuriman]

I occasionally do benches at 1.4-1.45v (= high temperatures along with high voltage), and so far have had no noticeable degradation of my chip.

 

[Burpo]

You didn't hurt anything. The chip throttles above 95c to protect itself. You didn't even get hot enough for that to happen.

 

[ShintaiDK]

The poor thing!

 

[KaRLiToS]

Nothing to worry about honestly.

 

[Dankk]

Thanks for the reassurance everyone. :$

 

[MongGrel]

You didn't hurt anything. The chip throttles above 95c to protect itself. You didn't even get hot enough for that to happen.

Thanks for the reassurance everyone. :$

Yeah, I beat on a few of mine myself, that thing wouldn't have been in trouble.

There is protection built into em to throttle down these days.

 

[2is]

You didn't hurt anything. The chip throttles above 95c to protect itself. You didn't even get hot enough for that to happen.

Throttling helps protect the CPU from burning up due to heat, it is not over-voltage protection. It is entirely possible to fry a CPU by pushing too much voltage even if you have the chip at sub freezing temperatures.

 

[greatnoob]

I think the older 100ish to 250nm chips ran at 1.5v like the old Pentiums. There is, of course, a massive difference between the chips from 2000 and those of 2014 but I doubt you've caused more than a couple of scratches worth of damage anyways.

 

[ClockHound]

Well...at least you didn't run benches, Dankk. I did. When I first started overclocking my Xeon 5660, I dyslexiated the voltage. Wanted to hit 4.4 stable, thought a bump up to 1.26 would do it. My fingers felt it wasn't enough, so set it at 1.62V. Rebooted.

Starting running LinX for a few minutes - noticed the temps had gone up quite a bit more than expected. Strange....glance over at CPUz and see 1.62V and scream at the screen. In a single fluid motion managed to leap out of the chair, knock over the case and pull the power cord.

Think I shortened my life more than the CPU's. Later learned from Burpo about using offset voltage. In small increments. With a single keystroke.

The Xeon is still running strong at quite a bit under 1.62V. And I'm still alive to tell the story. Happy ending all round.

 

[jhu]

Billions of dollars worth of equipment, centuries of man-hours spent for design/testing, just to produce that one CPU that you bought and installed... all for you to overvolt and potentially destroy your processor. You should feel ashamed.

 

[ninaholic37]

I think the older 100ish to 250nm chips ran at 1.5v like the old Pentiums.

Don't forget the 1946-1985 (ENIAC to some 80386s) that ran at 5V ... I think.

 

[soccerballtux]

Don't forget the 1946-1985 (ENIAC to some 80386s) that ran at 5V ... I think.

different semiconductor technology makes this a non-sequitor

 

[imported_Thorburn]

Don't forget the 1946-1985 (ENIAC to some 80386s) that ran at 5V ... I think.

The Intel 486DX2 was a 5V chip as well, the DX4 dropped that to 3.3V and you could get socket adapters to drop the voltage so you could use them in older boards.

With regards to the OP, it certainly isn't a good or wise thing to do, but if it was only for a short period of time then you may not experience any ill effects. I haven't really bothered overclocking since the Netburst days, but I had a Pentium D 940 in a Shuttle XPC system which I ran at 4GHz and the combination of high temperatures and voltage led to electron migration which reduced its stable operating frequency quite drastically over 6 months, eventually to the point where it wasn't stable at any speed. But that was sustained use with the system running 24/7 in those conditions.

I'd just keep using it but just bare in mind that in case of instability there is a higher than normal chance of the CPU being the root cause.