[Chris123NT]

This is a thread devoted to all of the hardware enthusiasts out there who seek to push their hardware to the absolute edge. So post your overclocks here! I'll start:

Core i7 920 D0 Batch 3845B260 @ 4.4 GHz

Attached File(s)

•  Stable.JPG (129.66K)
  Number of downloads: 45

[Eugene C]

Chris123NT, on 06 March 2010 - 01:13 AM, said:

This is a thread devoted to all of the hardware enthusiasts out there who seek to push their hardware to the absolute edge. So post your overclocks here! I'll start:

Core i7 920 D0 Batch 3845B260 @ 4.4 GHz

I think you won.

[Paul]

Wow, awesome OC there Chris

Glad to see you got it stable. I'm sure this is as far as you're gonna go though, as pushing it further is something you only want to do if you have a spare i7 920 lying around to take its place should you nuke it through overvolting.

[Chris123NT]

Paul, on 06 March 2010 - 03:56 AM, said:

Wow, awesome OC there Chris

Glad to see you got it stable. I'm sure this is as far as you're gonna go though, as pushing it further is something you only want to do if you have a spare i7 920 lying around to take its place should you nuke it through overvolting.

Pushing it any further for 24/7 usage would be iffy. I thought about gunning for 4.5 but I dunno if it's worth it lol

[Loebl]

Chris123NT, on 06 March 2010 - 06:29 AM, said:

Paul, on 06 March 2010 - 03:56 AM, said:

Wow, awesome OC there Chris

Glad to see you got it stable. I'm sure this is as far as you're gonna go though, as pushing it further is something you only want to do if you have a spare i7 920 lying around to take its place should you nuke it through overvolting.

Pushing it any further for 24/7 usage would be iffy. I thought about gunning for 4.5 but I dunno if it's worth it lol

depends... You could use the excess heat to power a pan and make your morning egg

[Link Tamake]

Offtopic Comment

My god Chris, I'm shocked your computer hasn't exploded yet.

[David]

Holy crap, LinX is a hell of a stress test though. There's no application out there that will ever put as much stress on your CPU as LinX does.

When running a Prime95 Blend test on my OC, my temps peak at about 55C. When running LinX, they peak at 85C. The only reason they don't go higher is because 85C is my emergency shutdown threshold

If you can run a few hours of P95 Blend, it's stable IMO. See if you can get to 4.5 with that.

[Ryan Price]

My old conservative ways...85C is just too hot in my opinion...even stressed. i remember when people would freak out over 55-60C!

[Chris123NT]

Huzzah! Will post more details later

 huzzah.png (52.95K)
Number of downloads: 31

[Patrick Laughner]

got it already huh.

[Chris123NT]

Patrick Laughner, on 28 April 2010 - 04:31 AM, said:

got it already huh.

Yeah, seems to be a damn good batch too, as most others are requiring well over 1.4v for 4.4 GHz, so definitely happy with it so far. It KILLS the 920 clock for clock.

[Mitchell]

Chris123NT, on 28 April 2010 - 04:33 AM, said:

Yeah, seems to be a damn good batch too, as most others are requiring well over 1.4v for 4.4 GHz, so definitely happy with it so far. It KILLS the 920 clock for clock.

So under that voltage wasn't possible for 4.4GHz then?

I know you were testing it at insanely low amounts last night.

[Chris123NT]

Mitchell, on 28 April 2010 - 06:51 AM, said:

Chris123NT, on 28 April 2010 - 04:33 AM, said:

Yeah, seems to be a damn good batch too, as most others are requiring well over 1.4v for 4.4 GHz, so definitely happy with it so far. It KILLS the 920 clock for clock.

So under that voltage wasn't possible for 4.4GHz then?

I know you were testing it at insanely low amounts last night.

Prime95 passed without issue at as low as 1.32 volts, but linx wouldn't run longer than 3 mins. Just goes to show why I use LinX over prime. The voltage is still pretty respectable though, most of the guys on the evga forums are breaking the 1.4v barrier to hit 4.4.

[Mitchell]

Chris123NT, on 28 April 2010 - 11:00 AM, said:

Prime95 passed without issue at as low as 1.32 volts, but linx wouldn't run longer than 3 mins. Just goes to show why I use LinX over prime. The voltage is still pretty respectable though, most of the guys on the evga forums are breaking the 1.4v barrier to hit 4.4.

Well you never know, Prime stable might just do it for 24/7 use but if not then I guess you could always play around with it more to get <1.388v at 4.4. The chip might just be capable of that lower Vcore.

I generally stress test on both Prime95 and LinX until I have an overclock 24/7 stable with both applications, although I do prefer LinX to Prime95 most of the time.

[Patrick Laughner]

You know, 1.4 ~ 1.45v is was most considered the absolute safe limit to the 45nm, So I'm guessing the 32nm might be a tad lower. I'd try not to go too far above 1.4v on it.

[Chris123NT]

Patrick Laughner, on 28 April 2010 - 01:15 PM, said:

You know, 1.4 ~ 1.45v is was most considered the absolute safe limit to the 45nm, So I'm guessing the 32nm might be a tad lower. I'd try not to go too far above 1.4v on it.

Intel's spec sheet says 1.4 so I'm trying to stay as close to that as possible, but there's guys running near 1.5v on good cooling and not having any problems.

These CPU's run COOL, my hottest core wasn't even hitting 60C last night under full load.

[Mitchell]

Patrick Laughner, on 28 April 2010 - 01:15 PM, said:

You know, 1.4 ~ 1.45v is was most considered the absolute safe limit to the 45nm, So I'm guessing the 32nm might be a tad lower. I'd try not to go too far above 1.4v on it.

Well according to Intel the max VID Voltage range for both 45nm and 32nm LGA1366 based Core i7's is 1.375v exactly. Personally I'd never run 45 or 32nm Intel chips above 1.42v for 24/7 use unless I was using a more extreme cooling solution such as Phase Change.

[Chris123NT]

Mitchell, on 28 April 2010 - 01:49 PM, said:

Patrick Laughner, on 28 April 2010 - 01:15 PM, said:

You know, 1.4 ~ 1.45v is was most considered the absolute safe limit to the 45nm, So I'm guessing the 32nm might be a tad lower. I'd try not to go too far above 1.4v on it.

Well according to Intel the max VID Voltage range for both 45nm and 32nm LGA1366 based Core i7's is 1.375v exactly. Personally I'd never run 45 or 32nm Intel chips above 1.42v for 24/7 use unless I was using a more extreme cooling solution such as Phase Change.

The spec sheet I saw for gulftown said 1.4, where did you see 1.375?

Anyway when I get home gonna try more tweaking. Apparently LOWERING the CPU PLL Vcore can get more stability, it's at 1.8 by default

[Mitchell]

Chris123NT, on 28 April 2010 - 02:00 PM, said:

The spec sheet I saw for gulftown said 1.4, where did you see 1.375?

I saw it on the Intel website under the VID voltage range for that chip.

http://ark.intel.com...t.aspx?id=47932

[Chris123NT]

Mitchell, on 28 April 2010 - 02:11 PM, said:

Chris123NT, on 28 April 2010 - 02:00 PM, said:

The spec sheet I saw for gulftown said 1.4, where did you see 1.375?

I saw it on the Intel website under the VID voltage range for that chip.

http://ark.intel.com...t.aspx?id=47932

Ah right, well I'm treating my own limit to 1.425 or so, it's crazy because the temps are so low, my radiator doesn't even get warm, and it definitely got warm when I would crank the 920 D0.