Sign in with
Sign up | Sign in
Your question

How much OC can YOUR Ivy Bridge take?

Tags:
Last response: in CPUs
Share
June 27, 2012 7:19:33 PM

I'm very curious as to how high your Ivy Bridge CPU's can go. Post your Ivy Bridge CPU's stable overclock speed!

-

I want to build a new system, and I'm wondering whether if I should buy the i7-2700K(SB has proven to be very OC friendly) or i7-3770K(IB seems to be not very OC friendly). Please and thank you for any information!

My current machine dinosaur:
Intel Socket 478 D875PBZ Motherboard
Pentium 4 EE 3.2GHz(OC'd to 3.5GHz, 875FSB, 2MB L3) CPU... Yes, 1 core...
2GB of RAM(OC'd to 438MHz from 400MHz)
Nvidia GeForce 7600GT(512MB, AGP) Graphics Card
120GB 7200RPM IDE HDD

^
So that's why I'm hoping to upgrade to a new computer to drop one of those processors in my next build. This computer crashes sometimes and reboots on me randomly, sometimes more than other times. Its time to move on.

More about : ivy bridge

a b à CPUs
June 27, 2012 8:05:19 PM

Hi :) 

You realise that ANY overclocking of a CPU can blow both it and the motherboard ??

Its ALWAYS a risk...

All the best Brett :) 
m
0
l
June 27, 2012 10:15:47 PM

I'm well aware of that, but remarks like those are things that kids in my school say about overclocking(no offense, they act like its pushing a self destruct button). I gave a specific, and legit reason as to why I'm asking this question(my dinosaur computer). Many people have had great results from overclocking, so I want to know from real people who've overclocked their Ivy Bridge CPU's as to how high they've taken their IB CPU's. Please don't try ruining a great topic, thank you.
m
0
l
Related resources
a b à CPUs
June 27, 2012 10:24:45 PM

As long as the temps are fine, OC'ing is not half as dangerous as people believe it is.
m
0
l
a c 188 à CPUs
June 27, 2012 10:25:16 PM

I have an Intel® Core™ i5-3570K that I have overclocked and reached a stable 4.6GHz while it was stable at this speed the temperatures on it were not what I would like them to be in a 24/7 overclocked system. So I ended up dropping it down to 4.2GHz to run all the time. At that speed you are still getting about the same performance from a 3rd generation Intel Core processor then you would have from a higher overclocked Intel Core i5-2500K. In the end on high end performance it ends up being about equal from between them. When you you add in things like PCI-E 3.0 and if you use them the better graphics the 3rd gen processors pull ahead.
m
0
l
June 27, 2012 10:56:47 PM

Should Ivy Bridge's thermal compound under the IHS be taken into consideration as to how long the CPU will last, or how long it will last under overclocked voltages & temperatures? Will a Sandy Bridge last longer than an Ivy Bridge under the same temperatures because the IHS and the CPU die are soldered together on a Sandy Bridge?
m
0
l
a b à CPUs
June 27, 2012 11:07:36 PM

My I7-3770k has run for 1.5 months at 4.5ghz @ 1.25v without a hiccup. Temps with a Corsair H80 peak at 82C in Intel Burn Test and reach a maximum of 80C in Prime95. I have gotten it stable up to 4.8ghz (1.32v) in games, but benchmark temps were unacceptably way up into the 90s and weren't showing signs of stopping their ascent. 4.5ghz is the sweet spot on my 3770k.

The other thing I might add is that absolutely nothing I could do would get that thing to boot with a base clock other than 100.0. If I managed to get into Windows it would promptly crash. This is on an Asus P8Z77-I Deluxe (a mini-ITX board) so I'm not really sure if that had anything to do with the board or whether my CPU was just Mr. Picky with its bclock.

Anyway what Christian mentioned was spot-on. If you are all about more ghz then the 2700k is really the OC champion on air. But Ivy has too many gaming-oriented benefits, better efficiency and better performance at the same clock to really go with SB.
m
0
l
a b à CPUs
June 27, 2012 11:30:03 PM

hyrule571 said:
I'm well aware of that, but remarks like those are things that kids in my school say about overclocking(no offense, they act like its pushing a self destruct button). I gave a specific, and legit reason as to why I'm asking this question(my dinosaur computer). Many people have had great results from overclocking, so I want to know from real people who've overclocked their Ivy Bridge CPU's as to how high they've taken their IB CPU's. Please don't try ruining a great topic, thank you.



Hi :) 

I along with everyone else is perfectly entitled to a view on this subject....

Plus I am an expert in what happens when it all goes wrong....

I own Computer shops and a Laptop repair company in the UK....

Several times a WEEK... we have to tell some unfortunate soul that they have blown their cpu or mobo or both, because they BELIEVED all the hype and "thought" there was no risk....

Well there is....and those people we have to give the bad news to are generally teenagers who are devasted that all their savings just went into the bin as it were....

All the best Brett :) 
m
0
l
a b à CPUs
June 27, 2012 11:32:10 PM

hyrule571 said:
Should Ivy Bridge's thermal compound under the IHS be taken into consideration as to how long the CPU will last, or how long it will last under overclocked voltages & temperatures? Will a Sandy Bridge last longer than an Ivy Bridge under the same temperatures because the IHS and the CPU die are soldered together on a Sandy Bridge?



Hi :) 

Makes no difference....

And overclocking always shortens the Cpu`s LIFE......and thats a fact.....

All the best Brett :) 
m
0
l
a b à CPUs
June 28, 2012 12:02:51 AM

Brett928S2 said:
And overclocking always shortens the Cpu`s LIFE......and thats a fact.....


Consider this. I've had an I7-920 D0 chip in an Asus P6TD board cooled by a Noctua NH-U12P for 2.5 years. Having spent a proverbial 'wad' on that cooler, I was curious just how much better it performed than the Intel stock HSF that came with the chip.

In a 21c a/c controlled environment, at total stock and fully-threaded the chip ran at its rated 2.66ghz, 1.2v vcore and peaked at 76c during an 8-hour Prime95 blend with the stock HSF. The same settings, same 8 hours of Prime95 blend yielded a 68c peak with the Noctua.

The stock heatsink faithfully kept the chip stable even to 3.5ghz @1.2v but temps peaked at 87c. The Noctua ran the same settings but peaked at 77c.

So let's get into it then: Being in the biz, you know the big killers of overclocked CPUs: #1: HEAT #2: Overvoltage #3: Cheap/inadequate motherboard VRM

So my OCed I7-920 runs 833mhz faster, same volts, and at the same temperature as it would @stock with the stock HSF by using far superior (and quieter I might add) cooling. It isn't being overvolted - this is what we call a "free" overclock. And finally, I am using a board that is explicitly designed to overclock socket-1366. It has 16+2 phase high-quality voltage regulation that is going to provide nice 'clean' power to my cpu. Where exactly am I decreasing my CPU or motherboard's life?

Brett - Rather than blaming the existence of overclocking on your customers' problems, maybe tell them that if they insist on overclocking, to follow certain guidelines like: 1) Use upgraded cooling! 2) Keep the vcore reasonable 3) Cheap motherboards offer overclocking too - but don't use those, use boards that can actually handle it. Just a suggestion.
m
0
l
June 28, 2012 12:12:35 AM

I may be a teenager going on my last year in high school Brett, but I could surprise you with how much I know with hardware and some networking(not to brag because you all, including you Brett, can probably smear me with more knowledge about this stuff). I'm well aware that both the motherboard and CPU have voltage, and temperature limitations. I've learned tons about computer hardware mainly by myself through diligent research and curiosity. I'm not going to argue with you because this topic doesn't have room for it, but what I will say is that I'm also well aware that a CPU's life is shortened when you overclock it. I'm not planning to overclock neither the SB, or IB CPU's over 4.5GHz. To me, if I were to overclock them over this speed I'd feel as if I'd have the computer for a significantly shorter time. Those teenagers who thought they knew what they were doing needed to consider that there's more to overclocking than just unknowingly raising the voltage. You also can't raise the FSB or Multiplier(depends on CPU) without incrementally raising them, not watching your temps. :non: 
m
0
l
a b à CPUs
June 28, 2012 12:13:19 AM

larkspur said:
Consider this. I've had an I7-920 D0 chip in an Asus P6TD board cooled by a Noctua NH-U12P for 2.5 years. Having spent a proverbial 'wad' on that cooler, I was curious just how much better it performed than the Intel stock HSF that came with the chip.

In a 21c a/c controlled environment, at total stock and fully-threaded the chip ran at its rated 2.66ghz, 1.2v vcore and peaked at 76c during an 8-hour Prime95 blend with the stock HSF. The same settings, same 8 hours of Prime95 blend yielded a 68c peak with the Noctua.

The stock heatsink faithfully kept the chip stable even to 3.5ghz @1.2v but temps peaked at 87c. The Noctua ran the same settings but peaked at 77c.

So let's get into it then: Being in the biz, you know the big killers of overclocked CPUs: #1: HEAT #2: Overvoltage #3: Cheap/inadequate motherboard VRM

So my OCed I7-920 runs 833mhz faster, same volts, and at the same temperature as it would @stock with the stock HSF by using far superior (and quieter I might add) cooling. It isn't being overvolted - this is what we call a "free" overclock. And finally, I am using a board that is explicitly designed to overclock socket-1366. It has 16+2 phase high-quality voltage regulation that is going to provide nice 'clean' power to my cpu. Where exactly am I decreasing my CPU or motherboard's life?

Brett - Rather than blaming the existence of overclocking on your customers' problems, maybe tell them that if they insist on overclocking, to follow certain guidelines like: 1) Use upgraded cooling! 2) Keep the vcore reasonable 3) Cheap motherboards offer overclocking too - but don't use those, use boards that can actually handle it. Just a suggestion.


Hi :) 

Your last paragraph is very true....

Most of the kids who come in with blown cpus/mobos tried overclocking on standard coolers and mobos and cases /fans...

But I never see them BEFORE they try it.....only usually when their machine will not boot...

These teenagers (usually) just "heard" about overclocking, and do it :( 

They are not bright enough to even come here and read some threads, which is why I usually give a warning in these threads.....

I realise that LOTS of people here can and do overclock successfully , but they are the brighter ones who realise there are limitations and actually know how to do it.......but LOTS of people also come into my shops who were unsuccessfull...

Regarding CPU hours (both unclocked and overclocked) there are plenty of technical articles around that say that overclocking shortens a cpu`s total life hours..... even Christian would agree with me on that lol :) 

All the best Brett :) 
m
0
l
a b à CPUs
June 28, 2012 12:14:48 AM

hyrule571 said:
I may be a teenager going on my last year in high school Brett, but I could surprise you with how much I know with hardware and some networking(not to brag because you all, including you Brett, can probably smear me with more knowledge about this stuff). I'm well aware that both the motherboard and CPU have voltage, and temperature limitations. I've learned tons about computer hardware mainly by myself through diligent research and curiosity. I'm not going to argue with you because this topic doesn't have room for it, but what I will say is that I'm also well aware that a CPU's life is shortened when you overclock it. I'm not planning to overclock neither the SB, or IB CPU's over 4.5GHz. To me, if I were to overclock them over this speed I'd feel as I'd have the computer for a significantly shorter time. Those teenagers who thought they knew what they were doing needed to consider that there's more to overclocking than just unknowingly raising the voltage. You also can't raise the FSB or Multiplier(depends on CPU) without incrementally raising them, not watching your temps. :non: 



Hi :) 

I totally agree with all you just said... :) 

But read my last post.... you are one of the brighter ones I mentioned...

All the best Brett :) 
m
0
l
a b à CPUs
June 28, 2012 1:35:36 AM

Brett928S2 said:
Regarding CPU hours (both unclocked and overclocked) there are plenty of technical articles around that say that overclocking shortens a cpu`s total life hours.....

Yes and they always point to heat and the amount and type of usage as the culprit. The i7-950 is the same silicon as i7-920 D0 with a locked multiplier of 23 instead of 20. If the 920's multiplier could be adjusted, most of them could run exactly the same as the 950. Are you saying the 950 has a shorter potential lifespan than the 920 because it's multiplier dictates a higher-clockspeed (assuming the extra heat was compensated for)? Don't get me wrong, I'm not someone who expects something for nothing, but I know that each chip is different and factory-set clockspeeds and voltages are very generic and conservative. Whether they still use a traditional 'binning' system, idk.

I suppose I will concede that with the locked multiplier, in order to achieve the same clocks as the 950 with a 920, the bclock (and every clock attached) is being cranked up and run out of spec. More stress on the memory-controller, ok I'll concede that. Never get something for nothing after-all.

But to get back to the CPUs the OP is considering. With proper cooling, there just isn't much of a problem with raising clock speeds on unlocked SB and IB chips by simply using the multiplier. I think that most people concede that there is risk to OCing, it certainly doesn't make it last any longer! But if you accept the risk, I think it's worth it. And it's a tough call whether to go SB or IB for an OC. I made the same call and am happy with Ivy (though I do often wonder whether I could hit 5ghz had I gone with a 2700k).
m
0
l
a c 185 à CPUs
June 28, 2012 1:59:54 AM

Brett928S2 said:
Hi :) 

Makes no difference....

And overclocking always shortens the Cpu`s LIFE......and thats a fact.....

All the best Brett :) 

This is unfortunately true. :cry: 

Even if your temperatures and voltages are low, overclocking regardless will shorten your CPU's life.
m
0
l
June 28, 2012 2:52:35 AM

Yeah, thank you Brett lol thanks for looking out though, honestly. Anyone got an Ivy Bridge CPU OC'd? List it here with the temperature reading too!
m
0
l
a c 185 à CPUs
June 28, 2012 3:14:16 AM

hyrule571 said:
List it here with the temperature reading too!

Too many factors. [:lutfij:2]
m
0
l
a b à CPUs
June 28, 2012 3:14:55 AM

amuffin said:
This is unfortunately true. :cry: 

Even if your temperatures and voltages are low, overclocking regardless will shorten your CPU's life.


Sorry for beating the dead horse and filling up hyrule's thread with this, but since the Muffin King is here I have to ask using a more current example this time: A i5-3450 vs. a i5-3550. Same silicon, higher multiplier. Assume the extra heat from the extra clockspeed of the 3550 is compensated for. By what you and Brett have said, the 3450 would outlast the 3550, right?

Can you point me to an article that would explain this better? I've scoured my sources and can't find anything that doesn't assume that an overclocked CPU will be hotter and/or use more vcore.
m
0
l
a c 185 à CPUs
June 28, 2012 4:15:25 AM

larkspur said:
Sorry for beating the dead horse and filling up hyrule's thread with this, but since the Muffin King is here I have to ask using a more current example this time: A i5-3450 vs. a i5-3550. Same silicon, higher multiplier. Assume the extra heat from the extra clockspeed of the 3550 is compensated for. By what you and Brett have said, the 3450 would outlast the 3550, right?

Can you point me to an article that would explain this better? I've scoured my sources and can't find anything that doesn't assume that an overclocked CPU will be hotter and/or use more vcore.

Theoretically, but remember not every CPU is the same.

EX: My 2500K can hit 4.5ghz stable with 1.236V, which is 0.016 more volts than stock.

This would be a good article to read :) 
http://www.tomshardware.com/reviews/ivy-bridge-overcloc...
m
0
l
a b à CPUs
June 28, 2012 4:33:54 AM

amuffin said:
Theoretically, but remember not every CPU is the same

Ok, but under the same theoretical example, you'd agree that if the 3450 had the stock cooler and the 3550 had some aftermarket cooler resulting in even a paltry 2c degree lower avg. temp, this would swing the theoretical potential lifespan of the 3550 higher than the 3450?

I'm just not seeing the danger of a chip that is simply running a higher multiplier at the same volts but is kept cooler through a better cooling solution than with a stock HSF configuration. Are we talking about the estimated maximum amount of switching the transistors will switch? Because that is far more dependent on how a user uses the CPU (e.g. folding@home 24/7) rather than the clockspeed.
m
0
l
!