The Reality of Haswell Overclocking - Results Poll

[Tweakin]

...I refuse to pay extra for something the manufacturer won't guarantee me (at least so far...)

I got my "k" for less than a Jackson over a non-k processor, and I think that was a good buy since it took my stock 3.4Ghz to 4.3Ghz, and across all cores within Intel's Voltage specs!. I think 900Mhz was a good buy as it is cheaper then the cost per Mhz for the stock core!

 

[OCGuy]

AMD did do a gauranteed OC chip recently.
Its called Fx 9590 & 9370.
AMD's move wasn't widely applaused because their processors are far behind Intel's in IPC.

So even a 5 Ghz Piledriver wasn't good enough to keep up with Intel. But they did deliever a gaurenteed OC chip.

Guaranteed OC? Any more info on that?

 

[rtsurfer]

Guaranteed OC? Any more info on that?

Well it runs 4.7 Ghz on all cores by default as you buy it.

And I havent heard of anyone having trouble pushing them to 5 Ghz & beyond.

Edit: It appears that it does require proper cooling though.

 

[BonzaiDuck]

Lebensraum. Tolerance. That's what makes for an OC-able core.

It's not just folklore: the die-shrinks are pushing things to a level where quantum physics kicks in. Partly because of the die-shrink, there is both better performance and lower power requirements. But unless they can reduce the power requirements disproportionately more relative to the lithography, I think there's less and less lebensraum.

"Slow death of Overclocking" one review article called it. That could either mean the processor OR the process and pastime.

 

[dragantoe]

Honestly I think upgrading from decent Sandy Bridge OC'd chip (like I did) doesn't make much sense either unless the PCIe 2.0 bus is keeping you up at night.

Honestly I can say the same with Westmere EP, I "upgraded" from a 3570k and many games got a huge boost in performance (especially emulators, that was an insane difference), and in 3dmark, I still get better physics scores than most 4770k's...

 

[tweakboy]

Only 4.1GHz on my 4770k :\

Haven't played around with the manual OC too much. Just used the auto OC. When I tried manually OCing, I got it up to 4.3GHz but every now and then I would get a restart. At 4.4GHz it would not boot up no matter what.

Nice m8

 

[Doreguul]

got my 4670k running at 4.5 for daily use at 1.29V. I can push it as high as 4.65 for benchmarks and such, but I don't like keeping the voltage over 1.3 (it just scares me no matter how cool it runs)

Validation from the other night after running OCCT

 

[Tweakin]

got my 4670k running at 4.5 for daily use at 1.29V. I can push it as high as 4.65 for benchmarks and such, but I don't like keeping the voltage over 1.3 (it just scares me no matter how cool it runs)

Validation from the other night after running OCCT

You have one of the good chips!:thumbsup:

 

[Doreguul]

yeah,' i've been lucky with this chip (I thought I screwed it up with the de-lidding). I also got a evga 780 classified hydro which is a beast with overclocking. I am running it with 1.32v @ 1411mhz on the core and 1840mhz on the memory. gpuz latest validation

 

[tential]

The people saying they got stable above 4.3 ghz without delidding are mostly lieing to themselves and others.

The general seething hatred against obvious stability tests is proof enough.

The same people hate running furmark with an unlimited bios on their GPUs as it shows their limits there as well.

Or it could be that people have stable overclocks for the applications they use and that's what counts for them?

Example: Your car has a 5 star safety crash rating. But does it have a 5 star safety crash rating vs a tank? Probably not. Yet does it really matter when you aren't crashing into tanks on the highway?

Same reasoning some people here have. Their OC is 5 star in the applications they use. Is it 5 star in a stress test that goes on for HOURS? Maybe not, but then again, stress tests aren't real world use or anything close to it.

 

[Durp]

Or it could be that people have stable overclocks for the applications they use and that's what counts for them?

Example: Your car has a 5 star safety crash rating. But does it have a 5 star safety crash rating vs a tank? Probably not. Yet does it really matter when you aren't crashing into tanks on the highway?

Same reasoning some people here have. Their OC is 5 star in the applications they use. Is it 5 star in a stress test that goes on for HOURS? Maybe not, but then again, stress tests aren't real world use or anything close to it.

At stock speeds processors are stable and pass any stress tests just fine. So yeah, "5 star safety when smashing into a tank"

But suddenly after overclocking, the definition of stable changes? I don't think so.

 

[Tristor]

Maybe I'm just paranoid, but I always go the extra length to ensure absolute rock solid stability of my system. I don't ever want to encounter something where I experience a crash or glitch because of an overclock, because I would find that frustrating. That's why I go to great lengths in cooling, take the time to work my way through the process slowly and find a point of stability that I'm happy with. Stress tests are an integral part of that because they represent the absolute maximum/worst case scenario for the chip. If it can handle that, it can handle anything.

I have no idea why people even argue over this. When you're overclocking don't you want the assurance your system can handle whatever is thrown at it gracefully? Using your car analogy, you're essentially advocating removing the seatbelts and airbags, because any "safety margin" for your overclock is eaten up if you can't pass a stress test at your settings. So what happens when you suddenly load up a game that's incredibly CPU intensive and it maxes all cores and now your system is crashing? Whoops, guess you should have tested better.

 

[Tweakin]

Maybe I'm just paranoid, but I always go the extra length to ensure absolute rock solid stability of my system. I don't ever want to encounter something where I experience a crash or glitch because of an overclock, because I would find that frustrating. That's why I go to great lengths in cooling, take the time to work my way through the process slowly and find a point of stability that I'm happy with. Stress tests are an integral part of that because they represent the absolute maximum/worst case scenario for the chip. If it can handle that, it can handle anything.

I have no idea why people even argue over this. When you're overclocking don't you want the assurance your system can handle whatever is thrown at it gracefully? Using your car analogy, you're essentially advocating removing the seatbelts and airbags, because any "safety margin" for your overclock is eaten up if you can't pass a stress test at your settings. So what happens when you suddenly load up a game that's incredibly CPU intensive and it maxes all cores and now your system is crashing? Whoops, guess you should have tested better.

Well said Tristor! I could be running my system at 4.4 but I would not consider it 24/7 stable as it will pass some stress tests for short durations, but not 24/7.

I have worked my way down from 4.4 to 4.2 and still dropping the voltage. For me this takes a great deal of time. I run IBT on 100 loops and AIDA64 runs for a couple of days and then I BOINC across everything for a week while I do normal work/surfing.

The ability for me to bench is meaningless...it's just an e-penis award. I use my system..