4930K @4.8GHz - Possible?

[utisz]

I've made a new pc this week. I'm pretty happy with it.
ASUS Rog Rampage IV Black Edition + 4930K + CorsairH90, seems like a good combo to me.

Anyway, I've successfully made a stable OC for my 4930K, 4,6GHz with 1.26 vcore. Depend,ng on what I see online, I think I am pretty lucky.

Right now max temp is around 66c, and there is plenty of flecibility there as well.

Depending on this information, do you think 4.8 is easily possible without eating up CPU life?

 

[ingtar33]

depends. remember with IB-E at about 1.25-1.30 the chip hits a temp wall, where temps start to skyrocket. you might be able to get a little more out of it, but don't be surprised if all of a sudden your temps are spiking out of control.

If you have a good cpu cooler you'll likely get a bit more out of the chip. there is no harm in trying it. Just know sustained temps in the +80C range is generally bad for intel cpus. the key word being sustained.

 

[Adroid]

I personally would quit while you are ahead. It would be a shame to burn a chip like that up lol.... 200 mhz isn't going to make that much of a difference.

And sorry, I can't personally speak to the 4930k, but for me 60s under load is good. Past 60s under load is pushing it. You will find plenty of opposing viewpoints on that subject, but again - just my opinion.

 

[utisz]

I tried pushing it to 4.8, could not succeed. Somehow, I can't get 4.6 with 1.26 vcore anymore... All I can get is 4.5 now. Any ideas what might've happened?

 

[Adroid]

What have you been stress testing with? I only ask because there is a chance that original clock wasn't stable to begin with.

Unless you ran too much voltage or temp, I wouldn't worry about it - Overclocking is not an exact science.

You should also check your event logger for whea errors. I added voltage to my chip until the whea errors stopped, hence my relatively low overclock. The thing runs stable with temps I am comfortable with though and it's FAST so what can I say.

 

[ingtar33]

I tried pushing it to 4.8, could not succeed. Somehow, I can't get 4.6 with 1.26 vcore anymore... All I can get is 4.5 now. Any ideas what might've happened?

probably just the temps in your room are hotter now then when you had it at 4.6

it's amazing what a few degrees will do.

 

[utisz]

The temps never exceed 66-ish - I'm getting BSOD's, possibly because of low power? However, that vuild used to work yesterday..

 

[ingtar33]

The temps never exceed 66-ish - I'm getting BSOD's, possibly because of low power? However, that vuild used to work yesterday..

i'm not talking about the cpu. lots of parts are affected by temps... ram, motherboards... .video cards. I'm saying a few degrees difference in the room might be all that it takes to cause a system to fail.

My fx8320 was like that when i was trying to reach 5ghz. One day it would pass a long stress test, the next it would crash while loading windows. It took sticking a 120mm fan over the VRMs on my motherboard to stabilize the overclock. The temps on the cpu weren't the issue. it was the temps of the vrms that was killing it. Depending on the time of the day the system would oscillate between stable and unstable. stable late at night and in the morning, unstable in the afternoon. the temps in the room didn't change all that much... maybe 2 or 3C... but that's all that it took.

 

[utisz]

So simply to say, it's not possible to "kill" the overclock potential of the pc by just playing around and trying to do more, like me?

 

[Adroid]

So simply to say, it's not possible to "kill" the overclock potential of the pc by just playing around and trying to do more, like me?

On the contrary, I have had couple different experiences with overclocking that are similar:

1) Sometimes I have reached a stable overclock, pushed farther, then revert to the overclock I had before and find it require more voltage to be stable than it did initially
2) Have a CPU that has been overclocked for years suddenly not like the overclock anymore, forcing me to reduce clock or add voltage.

Maybe coincidental? I can't say, but like I said before I don't think overclocking is an exact science. The good news is if your chip is holding at 4.5 with that low voltage you still have a better chip than most, not to mention pretty much the fastest (top 2%-5% at least) CPU in the world - so I wouldn't complain.

 

[Geek Jed]

Man, I got the same chip and have the same problem. I can't get the thing to be stable at 4.7, I want to go to 4.8.
A few people I read about had theirs up to 4.9 stable. Its a proven fact, liquid cooling really helps out if you overclock.

I had my 4930k at 4625MHz, 1.448v, just 5 hours ago then I went to do another Prime 95 stresser on it to make sure it holds and I got a BSoD.
Now its holding at 4.5 at same voltage.

 

[utisz]

So simply to say, it's not possible to "kill" the overclock potential of the pc by just playing around and trying to do more, like me?

On the contrary, I have had couple different experiences with overclocking that are similar:

1) Sometimes I have reached a stable overclock, pushed farther, then revert to the overclock I had before and find it require more voltage to be stable than it did initially
2) Have a CPU that has been overclocked for years suddenly not like the overclock anymore, forcing me to reduce clock or add voltage.

Maybe coincidental? I can't say, but like I said before I don't think overclocking is an exact science. The good news is if your chip is holding at 4.5 with that low voltage you still have a better chip than most, not to mention pretty much the fastest (top 2%-5% at least) CPU in the world - so I wouldn't complain.

Currently doing prime95 wfor the last ~85 minutes, it's holding 4.5GHz with 1.22vcore and around 64degrees(c)
Do you think this is a good result?

 

[Adroid]

So simply to say, it's not possible to "kill" the overclock potential of the pc by just playing around and trying to do more, like me?

On the contrary, I have had couple different experiences with overclocking that are similar:

1) Sometimes I have reached a stable overclock, pushed farther, then revert to the overclock I had before and find it require more voltage to be stable than it did initially
2) Have a CPU that has been overclocked for years suddenly not like the overclock anymore, forcing me to reduce clock or add voltage.

Maybe coincidental? I can't say, but like I said before I don't think overclocking is an exact science. The good news is if your chip is holding at 4.5 with that low voltage you still have a better chip than most, not to mention pretty much the fastest (top 2%-5% at least) CPU in the world - so I wouldn't complain.

Currently doing prime95 wfor the last ~85 minutes, it's holding 4.5GHz with 1.22vcore and around 64degrees(c)
Do you think this is a good result?

Insanely good. I think everyone who has the money to buy a 4930K wishes they had your chip.

 

[utisz]

Insanely good. I think everyone who has the money to buy a 4930K wishes they had your chip.

Oh!

 

[ingtar33]

So simply to say, it's not possible to "kill" the overclock potential of the pc by just playing around and trying to do more, like me?

On the contrary, I have had couple different experiences with overclocking that are similar:

1) Sometimes I have reached a stable overclock, pushed farther, then revert to the overclock I had before and find it require more voltage to be stable than it did initially
2) Have a CPU that has been overclocked for years suddenly not like the overclock anymore, forcing me to reduce clock or add voltage.

Maybe coincidental? I can't say, but like I said before I don't think overclocking is an exact science. The good news is if your chip is holding at 4.5 with that low voltage you still have a better chip than most, not to mention pretty much the fastest (top 2%-5% at least) CPU in the world - so I wouldn't complain.

well what you're describing is what happens when a cpu is dying. After time it takes more and more vcore to stabilize. The little know fact about intel cpus, is despite their higher reported operating temps (then AMD chips) they've always been prone to high temp burnout a lot easier. I've seen people (and been one of them) beat to death their AMDs, and do nothing more then kill the motherboard. I've killed several boards trying to overclock my amd chips... never managed to kill an AMD cpu. Yet i've also seen people (and done it myself) kill their intels without too much effort. That's why i posted earlier in this thread that sustained temps in the 80C range is inadvisable. It will kill an intel inside of a year.

Setting that aside, to the OPs question, no, it's unlikely you damaged your cpu unless you did something stupid like push the vcore too high too fast. That said stranger things have happened. I suspect your 4.6ghz overclock was never truly stable (like my original 5.0ghz overclocks weren't)... rather then scare yourself and think you hurt your chip, i expect the chip is fine and you're still trying to stabilize the overclock.


BTW: 4.5ghz on a 6 core i7-4930 is pretty good..

 

[utisz]

Setting that aside, to the OPs question, no, it's unlikely you damaged your cpu unless you did something stupid like push the vcore too high too fast. That said stranger things have happened. I suspect your 4.6ghz overclock was never truly stable (like my original 5.0ghz overclocks weren't)... rather then scare yourself and think you hurt your chip, i expect the chip is fine and you're still trying to stabilize the overclock.

BTW: 4.5ghz on a 6 core i7-4930 is pretty good..

Thank you for this additional information! Really helps, at least for how I feel
I've never tried more vcore than 1.35, never exceeded 73(c) in max. Only one time, the first time I set the vcore to "auto" and set the multiplier to 46, I've seen 1.55vcore and almost immediately restarted. I don't think that shouuld've harmed in ~2 minutes -- the temp was barely over 70..

 

[ingtar33]

Setting that aside, to the OPs question, no, it's unlikely you damaged your cpu unless you did something stupid like push the vcore too high too fast. That said stranger things have happened. I suspect your 4.6ghz overclock was never truly stable (like my original 5.0ghz overclocks weren't)... rather then scare yourself and think you hurt your chip, i expect the chip is fine and you're still trying to stabilize the overclock.

BTW: 4.5ghz on a 6 core i7-4930 is pretty good..

Thank you for this additional information! Really helps, at least for how I feel
I've never tried more vcore than 1.35, never exceeded 73(c) in max. Only one time, the first time I set the vcore to "auto" and set the multiplier to 46, I've seen 1.55vcore and almost immediately restarted. I don't think that shouuld've harmed in ~2 minutes -- the temp was barely over 70..

yeah. it should be fine. as i said... my origional 5.0ghz overclocks were all over the place. sometimes rock solid, then 2 hours later it would fail a stress test 30 seconds in. it would pass a 10 hour stress test then crash on windows bootup. as i said, sometimes all it takes is a slightly warmer temp in the room.

Its most likely you were never really stable at 4.6ghz.

still 4.5 is fantastic. and your voltages and temps are still low so it might be possible to stabilize higher.

 

[Adroid]

Setting that aside, to the OPs question, no, it's unlikely you damaged your cpu unless you did something stupid like push the vcore too high too fast. That said stranger things have happened. I suspect your 4.6ghz overclock was never truly stable (like my original 5.0ghz overclocks weren't)... rather then scare yourself and think you hurt your chip, i expect the chip is fine and you're still trying to stabilize the overclock.

BTW: 4.5ghz on a 6 core i7-4930 is pretty good..

Thank you for this additional information! Really helps, at least for how I feel
I've never tried more vcore than 1.35, never exceeded 73(c) in max. Only one time, the first time I set the vcore to "auto" and set the multiplier to 46, I've seen 1.55vcore and almost immediately restarted. I don't think that shouuld've harmed in ~2 minutes -- the temp was barely over 70..

yeah. it should be fine. as i said... my origional 5.0ghz overclocks were all over the place. sometimes rock solid, then 2 hours later it would fail a stress test 30 seconds in. it would pass a 10 hour stress test then crash on windows bootup. as i said, sometimes all it takes is a slightly warmer temp in the room.

Its most likely you were never really stable at 4.6ghz.

still 4.5 is fantastic. and your voltages and temps are still low so it might be possible to stabilize higher.

Yes, it might be possible to to stabilize higher, but the question is - at what cost?

Ivy bridge has an invisible voltage "wall". For me personally, that wall is at 4.3ghz. I CAN get it stable at higher, and I booted as high as 4.5-4.6, but with my cheapo air cooler, it's not worth the strain and high temps.

Is sounds like your "wall" may be at 4.5, in which case, there is a serious amount of diminishing returns here.. If you need to go from 1.22vcore to 1.4vcore for another 100-200mhz it really isn't worth it. Just my 2 cents.

 

[matsamas]

hallo everyone. I just want to share my opinion and experience like you do guys.
first and foremost don't be greedy. you have a really powerful chip and 4.5 is an amazing oc. Unfortunately your chip is that strong that pushing it 200mhz more will provide nothing more than a reduction on your chip lifetime.
I personally own a 3770k I choosed over 4770k due to better overclockability. I am using a z77x- ud5h mobo and a corsair h100i on push pull config. I reached 4.5ghz with a low 1.21v core. once a week I got a bosd with a code of 0x0000003A. after lot of research I found that my CPU needed more vcore. on 1.23 it was rock stable with 63 degrees after 11 hours on stress test. right after I went on ocing my sapphire hd 7970. I spent 1 week every single searching why my gpu can't overclock while being sure it wasn't defective.i could pass any gpu stress test but every games soon or later lead to bsod or PC freeze. guess what. There wasn't enough power on my CPU while both CPU and gpu where under heavy load. so I increased my v core to 1.26 and I am currently for 5 months on 4.5ghz with 68c under full load And my gpu is now 1200core from 925 and 1600mem from 1375.
well said all this long text but I was just like you.
ppl reached 4.8ghz with 1.3v core. well I said I can too. but it don't worth it. even my 3770k is blazing fast. everything runs absolutely smooth and stable. I do know
that 1-2 years it will require more v core to be stable at 4.5 cuz ocing sacrifice our chip lifetime in orded to increase performance. I prefer mine to 4.5 for 3-4 years than 4.8 for 1-2.do you agree?
none really wins when he is greedy.

 

[utisz]

You are right, thank you for the suggestions..
How many years do you think my chip will last in current settings? What defines the life?

 

[Davil]

You should make sure you have good fans in your case because with a liquid cooling block on your CPU the mosfets and vrms around the cpu socket won't get cooled like they normally would with a CPU fan. Also with a new processor it may take time before you find a truly stable set-up. Just about every processor I've ever bought (a lot) overclock differently from when I first bought them to after I had them for awhile. Sometimes better sometimes worse. Changing voltages isn't going to hurt it as long as you keep your temps down. I've seen people going as high as 1.6v for normal use and not having any problems. The curve of vcore to ghz past 4.5ghz on Haswells is pretty steep though. Eventually you have to start going .2v just to get another 100mhz out of it.

If I were you though I'd delid that thing and raise the voltage up to 1.5 and see what kind of speeds I could get because that is a pretty good chip you got.

 

[utisz]

The curve of vcore to ghz past 4.5ghz on Haswells is pretty steep though. Eventually you have to start going .2v just to get another 100mhz out of it.

Oh, I see. 4930K is ivy-e btw.

 

[utisz]

Depends on every thing really,How well you keep your system cool is one of them.And number two is keeping the dust out of case as much as possible.
You might get maybe 1 to 3 years at the most.

What is the lifetime normally (without oc) -
I've never overclocked, my cpu's last pretty longer in 7/24 use for rendering (render farm) purposes.. This is the first time I'm trying this actually and I wanted to get a little bir more out of what I have bought.

 

[Davil]

Yea I meant to include Ivy Bridge in that sentence, but they're both the same in that regard.

 

[matsamas]

good morning ppl. well the CPU life time differs in every chip. if you are at 4.5ghz with 1.22v core and temps like 70c under full load,I would give you at least 2 years stable but I have a reason to believe that your chip as my chip,are good ones due to easy overclockability. I personally expect mine to last in the exact same config for 3 years minimum.
I have the corsair h100i (push pull)on the top,3x120mm (1600rpm) front,1x120mm (2350rpm) rear. inside the case right next to my CPU I have ram air cooler. this is keeping my board,mostly the CPU and the NOrth bridge really cool. overall my PC is running on really low temps.