Intel Core i7-3960X (ES) Sandy Bridge-E Overclocked to 5.5 GHz

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Coolaler has posted overclocking results for a Core i7-3960X engineering sample in a pair of threads over at the coolaler forums. He demonstrates the 125 MHz base clock option on the SB-E platform. In addition, he overclocks his sample to 4800 MHz at 1.35V using traditional multiplier overclocking on air. In a second thread he overclocks the sample to 5500 MHz at 1.57V on dry ice cooling.


2.gif



11.gif



6.gif


source:

http://forum.coolaler.com/showthread.php?t=274080
(translate with google chrome)
 

Imouto

Golden Member
Jul 6, 2011
1,241
2
81
Google "Coolaler " before you run smack into the wall again...

Strange moon runes that some guys translate using google and belief as facts. Frigging useless leak site like the whole bunch on your sig.
 

Vesku

Diamond Member
Aug 25, 2005
3,743
28
86
Interesting timing on this info, sane OC not much different than 2xxxK. Nice to see on a monster die though.
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
crazy voltage for (essentially) an SB part. it wont last long like that.

Is it though? Its on dry-ice. Has anyone burned up their SB at that voltage and those temps? I'm really curious where the safe temps are for SB.
 

Lonbjerg

Diamond Member
Dec 6, 2009
4,419
0
0
Strange moon runes that some guys translate using google and belief as facts. Frigging useless leak site like the whole bunch on your sig.

Charlie guesses and fire broadsides...hopin shit will stick.

Coolaler OC's CPU's...that you compare the 2 are your problem/fallacy *shrugs*
 

LOL_Wut_Axel

Diamond Member
Mar 26, 2011
4,310
8
81
If you're not overvolting you're not trying am i right

For benchmark runs the most I'd personally run through SB is 1.4V. For 24/7 use, no more than 1.35V. If you're thinking of switching the CPU in a year or two, then I guess it's alright to run 1.4V 24/7 and 1.5V for 5GHz+ benchmark runs. But for three-five years, I think around 1.35V or a bit less is the best choice. :)
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
For benchmark runs the most I'd personally run through SB is 1.4V. For 24/7 use, no more than 1.35V. If you're thinking of switching the CPU in a year or two, then I guess it's alright to run 1.4V 24/7 and 1.5V for 5GHz+ benchmark runs. But for three-five years, I think around 1.35V or a bit less is the best choice. :)

These specific voltage's are based on...?
 

AdamK47

Lifer
Oct 9, 1999
15,842
3,628
136
Here we go with the SuperPI again.

I only joke about running CPUMark99, but this guy actually does it.

It's better than nothing though.
 

Borealis7

Platinum Member
Oct 19, 2006
2,901
205
106
how long is the 2011 platform expected to last? i know IVB will have 1155 chips, but is high-end IVB going to be on 2011?

if not then there is absolutely no point in getting an SB-E from a consumer standpoint.
luckily, SB-E is not meant for consumers! ;)
 

grkM3

Golden Member
Jul 29, 2011
1,407
0
0
Mmm, delicious degradation. But I guess if he only runs it at full load like that for a few minutes/hours, then it'll be okay.

I dono man,Iv been runnning my bench 2500k at 1.5 volts since the week they came out.Benched it for months at 1.55-1.57 volts at 5.3ghz

its still doing 5ghz and runs 4.7 at 1.44 with air cooling for months now.

I think vid will hit 1.37 volts on a stock sandy so your volts are on the safe side.

idle temps are 29-30 and hits 85c full avx linx load

There are a hell of alot of people pushings sandys over at xtreme systems and I dont see any threads of chips diying.

Iv heard of a few chips fried in the beggining becasue people didnt know how hard the digi vrm hits on the asus boards.

I have up to 200% current control on my board.If I set the offset @ 1.3 volts thinking its safe and have the cpu over current at 140 and llc at 75% the board will go from 1.3volts light load to 1.6volts linx load.

thats how some of those first chips went south
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
how long is the 2011 platform expected to last? i know IVB will have 1155 chips, but is high-end IVB going to be on 2011?

if not then there is absolutely no point in getting an SB-E from a consumer standpoint.
luckily, SB-E is not meant for consumers! ;)

Not meant for consumers :confused:

Ivy Bridge really isn't any more high end than SB is, it mostly facilitates higher clockspeeds. The real win is increased IGP and less power consumption.

I'm really interested in what the lead time between IVB and IVB-E will be - I don't think intel will continue the route that they did with sandy bridge. Its pretty crazy that SB was so strong that it eclipsed EE processors in so many applications - that has never happened with previous generation EE CPU's and I doubt intel intended for that to happen.
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
I dono man,Iv been runnning my bench 2500k at 1.5 volts since the week they came out.Benched it for months at 1.55-1.57 volts at 5.3ghz

its still doing 5ghz and runs 4.7 at 1.44 with air cooling for months now.

I think vid will hit 1.37 volts on a stock sandy so your volts are on the safe side.

idle temps are 29-30 and hits 85c full avx linx load

There are a hell of alot of people pushings sandys over at xtreme systems and I dont see any threads of chips diying.

Iv heard of a few chips fried in the beggining becasue people didnt know how hard the digi vrm hits on the asus boards.

I have up to 200% current control on my board.If I set the offset @ 1.3 volts thinking its safe and have the cpu over current at 140 and llc at 75% the board will go from 1.3volts light load to 1.6volts linx load.

thats how some of those first chips went south

Yeah, this is why I keep asking. At 4.7GHz my 2600K has a reported VID of 1.3961V.

Thus I'm curious about the voiced concerns regarding Vcc on SB's, are they born from caution/fear/paranoia or from data/specs/experience?

I personally don't know, hence my question.
 

LOL_Wut_Axel

Diamond Member
Mar 26, 2011
4,310
8
81
These specific voltage's are based on...?

On that I don't want heavy degradation and I'd want my CPU to last at least three years without having to raise voltage again, minimum. I wouldn't even consider selling a CPU that's been overvolted, so I'd pass it down to a secondary system once the time reaches.
 
Last edited:

Nemesis 1

Lifer
Dec 30, 2006
11,366
2
0
For benchmark runs the most I'd personally run through SB is 1.4V. For 24/7 use, no more than 1.35V. If you're thinking of switching the CPU in a year or two, then I guess it's alright to run 1.4V 24/7 and 1.5V for 5GHz+ benchmark runs. But for three-five years, I think around 1.35V or a bit less is the best choice. :)

You did see the 4.9ghz voltage did you not . 1.36 thats very safe . and one thing coolar does do most of the time in his previews is over volt . That is a fact.
 

Nemesis 1

Lifer
Dec 30, 2006
11,366
2
0
Yeah, this is why I keep asking. At 4.7GHz my 2600K has a reported VID of 1.3961V.

Thus I'm curious about the voiced concerns regarding Vcc on SB's, are they born from caution/fear/paranoia or from data/specs/experience?

I personally don't know, hence my question.

I don't no either IDC . But much of it comes from AND fanbois who love 1.5V on their systems . Thanks to SOI . Intels tri-gate with the quantuam well should prove very interesting when its released.