Intel Core i7-6700K ‘Skylake’ hits nearly 7GHz

[P4-630]

"A professional overclocker has set a new record for Intel Core i7-6700K microprocessor frequency. This time the new code-named “Skylake” processor hit nearly 7GHz, which is among the highest frequencies ever achieved by any Intel Core i-series microprocessors.

Chi-Kui Lam, a professional overclocker from Hong Kong-based HKEPC web-site, has managed to overclock an engineering sample of the Intel Core i7-6700K “Skylake-S” microprocessor to 6998.88MHz, or by whopping 74.97 per cent compared to its stock frequency. To hit nearly 7GHz, the overclocker had to disable three cores out of four and the Hyper-Threading technology. Voltage of the processor was increased from 1.2V to 1.888V."

http://www.kitguru.net/components/cpu/anton-shilov/intel-core-i7-6700k-skylake-hits-nearly-7ghz/

 

[Batou1986]

had to disable three cores out of four and the Hyper-Threading technology.

IMO doesn't count

 

[RejZoR]

If CPU isn't runnign full fledged (in this case, all 4 cores and HT), then it's pointless. Overlcock the CPU as it is designed or don't even bother.

 

[R-T-B]

If CPU isn't runnign full fledged (in this case, all 4 cores and HT), then it's pointless. Overlcock the CPU as it is designed or don't even bother.

For record counters it does. It's not like it would ever have been stable anyways... he's lucky he got the thing to boot long enough to take a screenshot.

It's a different world in record-breaking land.

 

[RejZoR]

It's nto exactly an achievement if you disbale 3/4's of the chip. I can also make a single cylinder car make 15k RPM. Which would totally fall apart with all 4 or more running...

 

[FordGT90Concept]

So...it has less performance than the same processor running with half of its cores activated.

 

[BiggieShady]

So...it has less performance than the same processor running with half of its cores activated.

To be exact, almost no performance at all, because it burns up in white smoke in the exact same moment you put some load on it ... what's impressive is it survived taking a screenshot

 

[RejZoR]

It's as pointless as building a car that has 10.000 HP. You can make it but you can't use it. What's the point? If you can make the highest overclock that can run SuperPi and be the fastest while doing it, then awesome. Just capturing a screenshot is a bit daft. It has no purpuse or function. A meaningless clock.

 

[Bill_Bright]

It is not pointless at all because it proves it can be done. They just had on the news today that scientist have grown first almost fully-formed human brain. Is it pointless since it is not "fully" formed and only the size of a pencil eraser? No - because it proves it can be done.

Now is this overclocking test practical? No.

I note also the CPU he used is an engineering sample, not a final production sample so that taints the test results, IMO.

What I really think is bogus is calling the guy a "professional overclocker". He is just a well funded enthusiast. And there is nothing wrong with that. But there is no special skill or advanced technical knowledge needed to achieve this that merits calling the guy a professional anything.

It's as pointless as building a car that has 10.000 HP. You can make it but you can't use it. What's the point?

That is not a good analogy. Ever heard of "Top Fuel Drag Racing"? They use it to make money - lots of money. And they are professionals because the builders must have advanced knowledge of physics, mechanics and fuels. And the driver who takes the car from 0 - 300 mph in 3.05 seconds is clearly skilled at keeping that screaming rocket running in a straight line - and on the ground.

It has no purpuse or function. A meaningless clock.

To you maybe. And frankly, to me too! I don't see the skill or fun in overclocking today. 15 - 20 years ago when I played around with overclocking, it was cool because it was something new, but it also involved running jumpers on motherboards to change timings. Like today, we had no schematics so we had to analyze the board to determine where to put the jumpers and you had to be skilled at modifying and soldering multi-layered PCBs so you didn't destroy the board in the process - just to get a few MHz increase on benchmarks.

But still, even then it was just for bragging rights. We could not "see" any performance gains in real-world use. We didn't earn any money from spectators or sponsors. There was no practical value at all because in the next month, faster CPUs came out anyway.

 

[RejZoR]

I respect overclockers who pump out insane clocks and run benchmarks on that. These proof of concept overclocks are just boring and pointless.

 

[Vayra86]

FWIW these records do tell something about the overclockability of chips. I don't recall such a record on any Ivy or Haswell chip, I think those only got as high as the lower regions of 6 Ghz.

 

[Bill_Bright]

These proof of concept overclocks are just boring and pointless.

Again, to you, maybe. But not others.

FWIW these records do tell something about the overclockability of chips.

I agree. They prove it can be done. But since the normal user cannot submerge their CPUs in liquid nitrogen, it is not a very practical experiment.

 

[Tatty_Two]

Personally I have no issues with disabling cores for extreme overclocks, if enthusiasts want to push core frequencies to the limit then fine where I do have an issue is then to call it "i7-6700K hits nearly 7gig", a 6700k has 4 cores and 8 threads

 

[HWTactics]

All record overclocks are done with the minimum amount of cores running; it's just the opposite of parallelism (many cores running at lower frequencies vs one core at a high frequency) and you can't have both -- I've seen a lot of these with only 2GB RAM running XP too --

I sure don't have the tools for an OC like that and a record is a record.

 

[Vayra86]

Again, to you, maybe. But not others.
I agree. They prove it can be done. But since the normal user cannot submerge their CPUs in liquid nitrogen, it is not a very practical experiment.

Well these results also hold true for less extreme overclocks. When extreme overclocking records are on the lower end, you can be pretty sure the chip is also not a great 'moderate' overclocker, and vice versa. Ivy is a perfect example of this compared to, for example Sandy which also had higher extreme results.

 

[Frag_Maniac]

This is nothing but a prime example of how ridiculous CPU OCing records are. It's bad enough they go only by clock speed with no concern for real world performance, but it's come to stripping them down to one core and no HT?

Might as well just fire up the old single core Pentiums and see how far you can clock those.

...since the normal user cannot submerge their CPUs in liquid nitrogen, it is not a very practical experiment.

"Submerge"? Are you even aware of how LN is used for such purposes?

 

[ZenZimZaliben]

I would have been way more impressed if it wasn't an ES Chip. This would have given me great hope for OC'ing the 6700k if it was the retail chip. Had it of been a Retail Chip, even though it would seem pointless to some it does set a baseline as to overclocking potential for the chip series.

 

[RealNeil]

I look forward to OCing the 6700K. (all four cores and running the two years long, "me" test)

 

[Aquinus]

But since the normal user cannot submerge their CPUs in liquid nitrogen, it is not a very practical experiment.

One does not simply "submerge" hardware in LN2. One pours LN2 into a container that sits upon a device. Submerging the entire thing in LN2 would not only be impractical, it could destroy the board and be dangerous to human health.

 

[Bill_Bright]

Wow! You folks sure are ready to pounce on others, huh? Submerge or sit upon, does it really matter? The point was, users are not going to have access to liquid nitrogen or know how to use it properly.

 

[jaggerwild]

I think it's funny how AMD holds most of the CPU over clock world records..........

 

[Bill_Bright]

I think it's funny how AMD holds most of the CPU over clock world records..........

Oh? I think that was true many years ago but not now.

http://www.cpubenchmark.net/overclocked_cpus.html

 

[RealNeil]

We used LN2 at work building airframes.
(wing pins)

 

[Bill_Bright]

I am sure you have a huge safety program in it's use, handling, storage, etc. too.

 

[Frag_Maniac]

Wow! You folks sure are ready to pounce on others, huh?

Uh, whatever Bill, just taking note of your oft very erroneous style of posting, but if it makes you feel better to imply we're all over you for no reason, go for it.

Submerge or sit upon, does it really matter?

It not only matter's, it's point worthy since you're often claiming to be correcting others on all things PC software and hardware. So yeah, it matters because as mentioned it's not only dangerous to user AND hardware to actually "submerge" such a part, it's also more than just impractical as mentioned, it's pretty much impossible.

The point was, users are not going to have access to liquid nitrogen or know how to use it properly.

Well we certainly know YOU don't know how to use it now, don't we? Suffice it to say such points are better made by those at least having a basic understanding of how LN is used for OCing.