• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

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

scotslad said:
doesnt the intel chipset software automatically install all the drivers anyway mate (once you download it) ?? I think it used to from what I remember, I could be wrong though, maybe the 2700K has a seperate graphics driver too, would be unusual to me though, hmmmm, interesting. :confused:
Click to expand...
no it doesn't . it seems to be a separate driver
 
Been playing with my newly built Gene-Z & 2700k this evening.

I need 1.36v for 4.8ghz & 1.44v for 5.0ghz

Temp @ 1.44 is around 76-77 so I think I'll probably stick with 4.8ghz.

Was hoping I would reach 5ghz on slightly less volts. My cpu is from the very first batch of OEM 2700k's.
 
Bluntwrapped said:
Been playing with my newly built Gene-Z & 2700k this evening.

I need 1.36v for 4.8ghz & 1.44v for 5.0ghz

Temp @ 1.44 is around 76-77 so I think I'll probably stick with 4.8ghz.

Was hoping I would reach 5ghz on slightly less volts. My cpu is from the very first batch of OEM 2700k's.
Click to expand...

whats the matter with 76-77 mate ? I would imagine you mean those temps are under severe stress which probably about 90% of users worldwide will never have thier cpu's running at anyway constantly, ya know what I mean ?

Or are you just meaning from the lifespan point of view of the cpu with the extra voltage ? I would guess you are, but to be honest I dont see what the fuss is becuase if your into upgrading etc then no doubt in a couple (maybe 3) years you will want to upgrade again anyway and even with the volts at 1.44 surely it would still have a lifespan way beyond 3 years ??? if not then these cpu's are rubbish and I wont invest in one :(

I have had my old i7 920 clocked at just slightly over 4G's for about 3 years now and its never ever had a problem, I know its a different cpu etc, but I would imagine that it kind of similar with lifespan and clocking etc, afterall I am pushing a cpu thats at stock 2.66-G's to approx 4 G's, you know what I am meaning mate ?

or maybe these new cpus are just rubbish, I dont know, I can only compare with what I have as I dont have a 2700K at present to compare it to, hmmmm, interesting all the same :)
 
Yep, after a minute or so of IBT @ 1.44v the temps jump up to the mid 70’s. I’m referring to Intel and OCuk’s overclocking guidelines (not exceeding 1.425v and keeping temps below 70oc). 1.4v puts out a lot of heat, it becomes near impossible to transfer the heat fast enough even with a custom loop. I’m trying to take everything into consideration for something that I can run 24/7 while having peace of mind.

I do believe these figures are a bit of a generalisation but I think I’m better playing safe than sorry. I may decide to push for 5ghz, I know the cpu is capable but again, I will have to consider the voltage bump and my temps under load.

The 2700k is 32nm while your i7 920 is 45nm. I’d feel much safer pushing 1.44v on a 920 than I would on my 2700k. I’m not saying 1.44v is definitely out of the question but I think in this case it’s worth taking the speculation of temps & volts into consideration. It is quite a bump to gain stability beyond 4.8ghz and as much as I’d rather settle for 5Ghz, it feels a little irresponsible to ignore the general consensus - bump the voltage higher than the recommended for the sake of a barely noticeable 200Mhz.

These CPU’s are great mate, it’s all just luck of the draw. As you can see in the other posts some folk have been really lucky 5Ghz @ 1.36v. I was drawn in by the first batch being “great clockers”. I realise now that this was also a massive generalisation. Also, there was never a statement to say these OEM chips hadn’t been tested and the good ones hadn’t been filtered out for the pre-built 5Ghz systems. If so, then the initial statement was pretty much BS.. seems to be a lot of that flying around at the moment!

I ran out of time last night but I will be doing more tinkering this evening. Will keep you updated.
 
scotslad said:
Or are you just meaning from the lifespan point of view of the cpu with the extra voltage ? I would guess you are, but to be honest I dont see what the fuss is becuase if your into upgrading etc then no doubt in a couple (maybe 3) years you will want to upgrade again anyway and even with the volts at 1.44 surely it would still have a lifespan way beyond 3 years ??? if not then these cpu's are rubbish and I wont invest in one :(
Click to expand...
running them at high voltage can kill them, also some users on this forum have said they chip has degraded after only 1 week.

you need to remember these chips are 32nm ,

also temps are important ocuk and intel say to keep it below 70c., and intel says 1.35v max

from 4.6ghz to 5ghz u won't notice any differece
 
Last edited:
gareth170 said:
running them at high voltage can kill them, also some users on this forum have said they chip has degraded after only 1 week.

you need to remember these chips are 32nm ,

also temps are important ocuk and intel say to keep it below 70c., and intel says 1.35v max

from 4.6ghz to 5ghz u won't notice any differece
Click to expand...

ohh I see :)

So I am guessing but are you saying the old i7 920 like I have just now was and is infact a much more clockable cpu when it comes to volts and temps as compared to these new types, yea ? ;)
 
scotslad said:
ohh I see :)

So I am guessing but are you saying the old i7 920 like I have just now was and is infact a much more clockable cpu when it comes to volts and temps as compared to these new types, yea ? ;)
Click to expand...
your i7 920 is 45nm hence why higher voltage doesn't effect them as much but can run hotter + use up more power.

usely smaller the die the less heat it gives out/ less power/ lower voltage or less voltage
 
Last edited:
so what about the i7-3930K and the ivy, are they going to be as rubbish in comparison to the old i7 920 too ??? (i mean in the volts and temps etc).
 
scotslad said:
so what about the i7-3930K and the ivy, are they going to be as rubbish in comparison to the old i7 920 too ??? (i mean in the volts and temps etc).
Click to expand...
just because the safety voltage / temps limits are lower then the 45nm doesn't mean its rubbish.

you'll be able to overclock a 32nm chip higher than a 45nm chip with most likely less voltage..

a Sandy Bridge-E and Ivy Bridge is 22nm so the safety limits are most likely slightly lower than a 32nm..
 
gareth170 said:
just because the safety voltage / temps limits are lower then the 45nm doesn't mean its rubbish.

you'll be able to overclock a 32nm chip higher than a 45nm chip with most likely less voltage..

a Sandy Bridge-E and Ivy Bridge is 22nm so the safety limits are most likely slightly lower than a 32nm..
Click to expand...

I see mate, thanks, interesting stuff, hmmmm. ;)
 
I've got my OEM 2700K up to 5 GHz so far, at 1.43V. It's under chilled water (chilled to 12C), and the maximum core temp during 5 runs of IBT 2.52 (with AVX) at max stress is 56C.

I'll be pushing it a litle more in a few days, when I redo the loop and add the gpu I'm waiting on (then I can do some more extensive benchmarking). I'm going to follow the guide at the ROG forums for my board as well and play around with bclk, skew, etc. to see what (if any) differnce they make (I'm especially interested to see if adjusting the skew will affect core temps and min Vcore for stable operation).

One thing I did find interesting at 4.8 GHz was that when chilled, as well as knocking about 30C off the max core temp, I was able to drop Vcore by 0.02V (from 1.38V to 1.36V) and pass 5 runs of IBT.

I'm also considering buying a TEC block (as well as using the chiller) to see how that affects performance.

 
Last edited:
debs3759 said:
I've got my OEM 2700K up to 5 GHz so far, at 1.43V. It's under chilled water (chilled to 12C), and the maximum core temp during 5 runs of IBT 2.52 (with AVX) at max stress is 56C.

I'll be pushing it a litle more in a few days, when I redo the loop and add the gpu I'm waiting on (then I can do some more extensive benchmarking). I'm going to follow the guide at the ROG forums for my board as well and play around with bclk, skew, etc. to see what (if any) differnce they make (I'm especially interested to see if adjusting the skew will affect core temps and min Vcore for stable operation).

One thing I did find interesting at 4.8 GHz was that when chilled, as well as knocking about 30C off the max core temp, I was able to drop Vcore by 0.02V (from 1.38V to 1.36V) and pass 5 runs of IBT.

I'm also considering buying a TEC block (as well as using the chiller) to see how that affects performance.

Click to expand...

Interesting stuff. Did you run 1.43v without the chiller? What were your temps like before, or was that 30c lower than the higher voltage?
 
Last edited:
Bluntwrapped said:
Interesting stuff. Did you run 1.43v without the chiller? What were your temps like before, or was that 30c lower than the higher voltage?
Click to expand...

At 4.8 GHz on air my max temp was 76C when testing with IBT (max). Under water the max was 53 before adjusting the voltage, 51 after (so only 25 less, not the 30 I had thought). I didn't try any higher under air as I didn't want to risk the high temps at 1.4V and higher.

The variance between different cores was as high as 10C when I first tested under water. I had already set it to 5 GHz before I reapplied the TIM (it was even across the face of the processor, but I suspected that to be the cause of the wide variance). After reapplying TIM, the variance dropped to 5C, which I'm much happier with.

I'm now a little impatient for the gpu and a few more wc bits to arrive so I can redo the loop (I'm experimenting to get an idea of what will be best on my dual-chip build for next year) and play a little more. With the gpu installed, I will know that any stable changes to bclk and skew will remain stable :)
 
I built a new system this weekend with a OEM 2700K, so far it's running at 4.6 @ 1.30V. It's passed 6 hours of LINX, with temps at around 60c - 65c with a corsair H100. I think I will apply my own thermal paste this week as I have stuck with the standard stuff which comes on the corsair unit for now.

Having read this thread I'm quite sure 4.6 is very good for the voltage I'm running, once I've lowered my temps a bit more I will be pushing for at least 4.8 as long as my motherboard does not hold me back.

I'm new to sandybridge overclocking, but does any one recommend using the BLCK for small increments on top of the turbo multiplier IE a BLCK of 102.2 ??
 
BCLK should really only be tweaked if you are looking for maximum performance and have a decent custom water loop (or better). As it affects the speed of everything on the motherboard and can create more instability, you are best to stick to multiplier clocking for most overclocking.

Having said that, I will be tweaking BCLK, skew and other settings soon, just gotta finish my build (the GTX 570 arrived today and the last few parts to finalise my cooling loop are scheduled to arrive tomorrow).
 
You must log in or register to reply here.
Back
Top Bottom