Those of you with OcUK OEM 2700K Official OC Results thread!!

Gibbo · 29 Oct 2011 at 23:15

Hi there

Well as sales of our OEM 2700k is through the roof, no doubt because the word is out they are all first batch silicon and should prove best clockers, I thought a thread should be made to post overclock results. So I shall start:-

Max prime stable speed (minimum 1hr) - 5000MHz
Max booting into windows speed - 5400MHz
Motherboard - Asus Maximum Formula with BIOS set to AUTO mainly, just multiplier and vcore used.
Cooler - OcUK H2Flo with two fans
Memory - Kingston Genesis 8GB Kit

Prime stable at 5000MHz with 1.41-1.42v

Unstable boot into windows at 5400MHz with 1.55v

Be good to see others results in here from 2700k owners and please state if your CPU was an OcUK OEM one.

easyrider · 29 Oct 2011 at 23:18

Gibbo me ol mate send me one through the post and ill clock one for ya!

We go way back!

GoogalyMoogaly · 29 Oct 2011 at 23:18

Oh come on!
I'm trying to resist getting a SB and waiting for SB-E or IvyBridge. Seeing people getting big clock on these is not going to help!
Plus there's the pressure of the whole "first batch" thing!
Damn you Gibbo!

Gibbo · 29 Oct 2011 at 23:21

GoogalyMoogaly said:
Oh come on!
I'm trying to resist getting a SB and waiting for SB-E or IvyBridge. Seeing people getting big clock on these is not going to help!
Plus there's the pressure of the whole "first batch" thing!
Damn you Gibbo!

Its why I purchased 500 out of the 700 of the first batch, because I knew their would be many keen people who'd want a good clocking first batch CPU. Worse case is they don't sell and well we just build systems with them, but I'd rather customers got them and clocked them to great levels.

As soon as the first 500 sell out I shall let you all know as they won't then be first batch any longer, but we've only sold 30 so far so plenty left to go.

GoogalyMoogaly · 30 Oct 2011 at 01:02

Now the thing that concerns me, and I may just be being paranoid, is that since these are OEM chips that OcUK will have tried and tested them before hand and filtered off the really good overclockers for OcUK builds, staff and friends. Which leaves me/us with the rejected overclockers.

Other than that I am getting very close to saying to hell with waiting for IvyBridge (do we have an ETA on these yet?) and getting one of these and then seeing if IvyBridge is a worthy upgrade.

Hotwired · 30 Oct 2011 at 02:07

GoogalyMoogaly said:
Now the thing that concerns me, and I may just be being paranoid, is that since these are OEM chips that OcUK will have tried and tested them before hand and filtered off the really good overclockers for OcUK builds, staff and friends. Which leaves me/us with the rejected overclockers.

You really think that someone is going to test 500 chips?

Or that Esnet is made of obsessive computer nerds who are desperate to raid the new batch to keep up with the overclocking scene?

Yeah paranoid

gurusan · 30 Oct 2011 at 02:29

GoogalyMoogaly said:
Now the thing that concerns me, and I may just be being paranoid, is that since these are OEM chips that OcUK will have tried and tested them before hand and filtered off the really good overclockers for OcUK builds, staff and friends. Which leaves me/us with the rejected overclockers.

no

Gibbo · 30 Oct 2011 at 07:13

GoogalyMoogaly said:
Now the thing that concerns me, and I may just be being paranoid, is that since these are OEM chips that OcUK will have tried and tested them before hand and filtered off the really good overclockers for OcUK builds, staff and friends. Which leaves me/us with the rejected overclockers.

Other than that I am getting very close to saying to hell with waiting for IvyBridge (do we have an ETA on these yet?) and getting one of these and then seeing if IvyBridge is a worthy upgrade.

We've tested two!

One I took, it did 5GHz easy.
Another for a system build which ran at it's 4.80Ghz tested speed no issue, higher was not tried as it was a system build.

So out of 500 we've used 2, we don't pretest CPUs if we did well we'd be charging much higher price.

Plus with this batch there is little point we believe most will do 5ghz quite easily.

Purgatory · 30 Oct 2011 at 07:42

Nice tests. Just a a quick question I thought this applied to all Sandy Bridge chips ?

***OVERCLOCKING GUIDELINES***

- Do not exceed 1.35-1.38v core voltage, doing so could limit lifespan of the CPU

Now I see the site has changed it to (for 2600k and 2700k):-

***OVERCLOCKING GUIDELINES***

- Do not exceed 1.425-1.450v core voltage, doing so could limit lifespan of the CPU

Is that not way off intel's safe overclocking specifications ? Also didn't people before damage chips at voltages of 1.4v + core voltage ?

I see you changed the 2600K overclock guidelines to the same as the 2700k, but the 2500K is still on the old safer overclock guideline settings.

Has something changed recently with the 2600k and 2700k chips to allow such an increase in Vcore to be classed as safe now ? If that is the case so would it not be the same for the 2500k chps ? Also nothing has really changed on the 2700k from the 2600k apart from it's got a 100mhz boost in speed same stepping etc etc. Yes I do understand the principle of first batch silicon and can see some benefits to them chips for overclockers.

Gibbo · 30 Oct 2011 at 11:02

Purgatory said:
Nice tests. Just a a quick question I thought this applied to all Sandy Bridge chips ?

***OVERCLOCKING GUIDELINES***

- Do not exceed 1.35-1.38v core voltage, doing so could limit lifespan of the CPU

Now I see the site has changed it to (for 2600k and 2700k):-

***OVERCLOCKING GUIDELINES***

- Do not exceed 1.425-1.450v core voltage, doing so could limit lifespan of the CPU

Is that not way off intel's safe overclocking specifications ? Also didn't people before damage chips at voltages of 1.4v + core voltage ?

I see you changed the 2600K overclock guidelines to the same as the 2700k, but the 2500K is still on the old safer overclock guideline settings.

Has something changed recently with the 2600k and 2700k chips to allow such an increase in Vcore to be classed as safe now ? If that is the case so would it not be the same for the 2500k chps ? Also nothing has really changed on the 2700k from the 2600k apart from it's got a 100mhz boost in speed same stepping etc etc. Yes I do understand the principle of first batch silicon and can see some benefits to them chips for overclockers.

Yes Intel tech documents show a range of upto 1.425v as safe.

We deem upto 1.45v is safe, any higher though is unknown territory.

Gibbo · 30 Oct 2011 at 11:03

I've just taken a look at one of the CPU's for stepping info, here it is:-

SR0DG stepping
MALAY
L114B977

All of our OEM are this MALAY SR0DG stepping and so far all seem to do 5GHz with a breeze, look forward to see more results from customers.

mmj_uk · 30 Oct 2011 at 11:04

Have you tested any retail samples Gibbo?

Gibbo · 30 Oct 2011 at 11:26

mmj_uk said:
Have you tested any retail samples Gibbo?

Nope and we have small numbers as such we can't guarantee a specific batch so we won't be testing those.

With the OEM we can guarantee the same batch due to buying 500 of them.

Greg0986 · 30 Oct 2011 at 13:14

Gibbo said:
Yes Intel tech documents show a range of upto 1.425v as safe.

We deem upto 1.45v is safe, any higher though is unknown territory.

Thanks for the update!

I have been holding back because you have said 1.38v is maximum safe voltage

. 5GHz here I come!

GoogalyMoogaly · 30 Oct 2011 at 14:45

Hotwired said:
You really think that someone is going to test 500 chips?

Or that Esnet is made of obsessive computer nerds who are desperate to raid the new batch to keep up with the overclocking scene?

Yeah paranoid

gurusan said:
no

Gibbo said:
We've tested two!

One I took, it did 5GHz easy.
Another for a system build which ran at it's 4.80Ghz tested speed no issue, higher was not tried as it was a system build.

So out of 500 we've used 2, we don't pretest CPUs if we did well we'd be charging much higher price.

Plus with this batch there is little point we believe most will do 5ghz quite easily.

OK fair enough. I'm just sure in the past I've seen people (in build logs) say that they have a cherry picked CPU thanks to "OcUK employee X" or similar. Admittedly back in the i7 9xx days.

Still, good to know. It's not helping me resist a SB upgrade though!

mmj_uk · 30 Oct 2011 at 15:06

So has anyone apart from Gibbo and Boostjunky got one to 5ghz? Zoomee in another thread seems to only have his at 4.6ghz.

I generally get a bit suspicious when a shop employee starts hyping something that they have a lot of to shift.

Geeman1979 · 30 Oct 2011 at 15:20

Gibbo said:
Hi there

Well as sales of our OEM 2700k is through the roof, no doubt because the word is out they are all first batch silicon and should prove best clockers, I thought a thread should be made to post overclock results. So I shall start:-

Max prime stable speed (minimum 1hr) - 5000MHz
Max booting into windows speed - 5400MHz
Motherboard - Asus Maximum Formula with BIOS set to AUTO mainly, just multiplier and vcore used.
Cooler - OcUK H2Flo with two fans
Memory - Kingston Genesis 8GB Kit

Prime stable at 5000MHz with 1.41-1.42v

Unstable boot into windows at 5400MHz with 1.55v

Be good to see others results in here from 2700k owners and please state if your CPU was an OcUK OEM one.

Is it safe to run these CPU,s with INTERNAL CPU PLL OVERVOLTAGE enabled 24/7 for your overclock? Ive never really seen this answered.

evilpenguin · 30 Oct 2011 at 15:34

GoogalyMoogaly said:
OK fair enough. I'm just sure in the past I've seen people (in build logs) say that they have a cherry picked CPU thanks to "OcUK employee X" or similar

if a chip doesn't clock we normally replace it with one which does, although one not clocking is rare. we don't play favourites so i think people saying they have a cherry picked chip might be getting confused with "properly tested and clocked"

Gibbo · 30 Oct 2011 at 16:07

geeman1979 said:
Is it safe to run these CPU,s with INTERNAL CPU PLL OVERVOLTAGE enabled 24/7 for your overclock? Ive never really seen this answered.

I just set the CPU at 1.41v so it runs at that always even when not loaded as I'm used to the old methods of clocking.

zoomee · 30 Oct 2011 at 16:08

mmj_uk said:
So has anyone apart from Gibbo and Boostjunky got one to 5ghz? Zoomee in another thread seems to only have his at 4.6ghz.

I generally get a bit suspicious when a shop employee starts hyping something that they have a lot of to shift.

Not that I doubt the skills of the Ocuk team - but I personally prefer 100% stability over an extra couple of hundred mhz. Bear in mind though I am also running DDR3-1866 ram @ 1.65v (stock settings of 9-11-9-27-1T) and this is a retail 2700k from another retailor (early batch I assume as I got it on wednesday).

Prime95 is great stability tool for testing memory, subsystem and CPU - but does not stress the CPU to its fullest - Best test for that is IBT 2.51 but to set to MAXIMUM and also to set to 8 Threads (auto setting doesnt do this properly). This ensures 200% stability! As evidenced below - My prime95 temps are quite lower than my IBT Max test results:

I'm using an ASRock Z68 Extreme 4 Gen3 board with a simple AUTO OC Profile of 4.6Ghz - it runs at approx 1.28v

. Saying that I've also got custom watercooling 120.3 + 2x5850's+DDC+XSPC Res etc watercooled with EK blocks. I briefly tried 4.8 and 5 Ghz with approx 1.4v - both failed (and that was after clocking my memory down to DDR3-1600 but also needed PLL overvoltage enabling which I wasnt comfortable with!).

I know that using higher voltage on ram affects stability of the CPU at higher clocks so this could be a contributing factor as to why its not doing so well higher than 4.6 - I've always said its time to stop oc-ing once the scaling factor doesn't respond very well. i.e. [email protected] = perfect. [email protected]=too much - look at the jump in voltage required.

PS WARNING - ZOOMEE IS TEH SANDYBRIDGE NOOB