PDA

View Full Version : 1.475v too high for a 65nm CPU???


sillyeagle
01-27-08, 11:06 PM
What do you think? Toms says they would never recommend going as high as 1.475, but people here have assured me I can easily run 1.5v. Thats a pretty big difference.

http://www.tomshardware.com/2007/11/08/dual_vs_quad/page14.html

To achieve an overclock of 934 MHz, we had to increase the core voltage by an amount that becomes dangerous for the CPU. In the BIOS, we had to set a voltage tweak of 0.11250 Volts, for a core voltage of 1.46250.

We were even able to overclock our E6750 by 1.04 GHz up to 3.7 GHz. At this speed, the FSB runs at 463 MHz (1852 QDR). Although we raised the core voltage to 1.4750.

At this point, we decided to discontinue our overclocking tests, since the risk of damaging our CPU simply became too great for our liking. We can't really recommend running a 65-nm processor at a voltage this high anyway.

Uberpwnage
01-27-08, 11:12 PM
1.55v is the max "safe" voltage for 65nm according to Intel AFAIK, so most people recommend not going past 1.5v. You should be fine at 1.5v, everything I have seen up to this point has stayed concurrent with that. Idk why Tom's seems to think 1.475 is so horrible...maybe it's because they're using a Zalman cooler, so on a TRUE for example, 1.5 would be fine.

However, 45nm is a different story, as max "safe" according to Intel is 1.45v so 1.4v is the recommended.

Buckeye
01-27-08, 11:40 PM
I was able to get mine to 3.8ghz with 1.5v but the Freezone was not able to keep up with the temps for very long. Pretty crappy cooling unit when you start getting the OC high.

AthlonXP1800
01-28-08, 12:17 AM
1.55v is the max "safe" voltage for 65nm according to Intel AFAIK, so most people recommend not going past 1.5v. You should be fine at 1.5v, everything I have seen up to this point has stayed concurrent with that. Idk why Tom's seems to think 1.475 is so horrible...maybe it's because they're using a Zalman cooler, so on a TRUE for example, 1.5 would be fine.

However, 45nm is a different story, as max "safe" according to Intel is 1.45v so 1.4v is the recommended.

Hmmm a user from xtremesystems got E8200 overclocked to 4.8GHz rocked stable on 600MHz FSB at 1.488v with freezer 7 pro CPU cooler on air here (http://www.xtremesystems.org/forums/showthread.php?t=174820).

WeReWoLf
01-28-08, 12:42 AM
To hit 3.6 on my E6600, i have to hit 1.6v... Thankfully i only ever do that when benching, and in Winter. The rest of the time i have it at 3.2 on 1.4v

Uberpwnage
01-28-08, 01:53 AM
I'm pretty sure Intel's definition of max "safe" voltage is for a user running on Air cooling, 24/7. With better cooling, it's reasonably safe to raise the voltage above Intel's spec. However, every chip has a breaking point, no matter how well it is cooled. This bar has been lowered across the board in the transition from 65nm to 45nm.

Hmmm a user from xtremesystems got E8200 overclocked to 4.8GHz rocked stable on 600MHz FSB at 1.488v with freezer 7 pro CPU cooler on air here (http://www.xtremesystems.org/forums/showthread.php?t=174820).

Most of the overclocks you see at xtremesystems aren't for 24/7 usage. They are simply achieved for benching, or to determine the maximum capability of a given CPU or mobo. Sure, you can pump alot of volts into a 45nm on air, but to run it like that all the time will burn out the chip before long. Here (http://www.xtremesystems.org/forums/showthread.php?t=170358) is an example of how acceptable 65nm voltage (even with "sufficient" cooling) can kill a 45nm. Obviously the voltage used here was drastically higher, but so was the cooling, and it would have never burned out a comparable 65nm chip in such a short period of time.

Dazz
01-28-08, 02:08 PM
I run my E4300 @ 1.66v 3.5GHz

stncttr908
01-28-08, 03:17 PM
I run my E4300 @ 1.66v 3.5GHz
Good to know. I've only pushed mine to 1.5V to reach 3.5GHz (see sig), but never thought it to be safe. Oh well, I'm going 45nm in the near future so I may as well torture this thing.

AthlonXP1800
01-28-08, 06:40 PM
I run my E4300 @ 1.66v 3.5GHz

Your PC clocked 3.5GHz running 24/7 and since how long?

Uberpwnage
01-28-08, 06:45 PM
I run my E4300 @ 1.66v 3.5GHz

at that voltage, hopefully you're atleast on decent water cooling...

stncttr908
01-28-08, 07:01 PM
Your PC clocked 3.5GHz running 24/7 and since how long?
My everyday clock is 3.2. I only had it up at 3.5GHz and 1.5V for maybe an hour or so, running through Prime95 and Orthos. I might kick it up after reading that.

Slammin
01-28-08, 07:33 PM
I run my E4300 @ 1.66v 3.5GHz



I thought I read somewhere that the cpu would not even take more that 1.55v or something like that.

My CPU absolutely hates anything above 1.53v, regardless of speed.

KasuCode
01-28-08, 08:11 PM
9650 @ 4.4Ghz @ 1.56v 4.2Ghz @ 1.50 24/7 but on a very good water setup.

Slammin
01-28-08, 11:10 PM
Rockin!!

Dazz
01-29-08, 01:49 PM
I run at 3.5GHz but have watercolling, i did used to use 1.66v @ 3.4GHz with aircooling but was not stable due to the PWR running hot. But they are much cooler now i have watercolling which has allowed me to get stable to 3.5GHz. I run my PC about 8+ hours aday but is rearly left on for more then 24 hours.

sillyeagle
02-01-08, 03:16 AM
My E6750 didn't end up being the best OCer. I can do 3.4Ghz @ 1.47v or less, but 3.5GHz requires 1.5v and can't do much more than 30 minutes of Orthos small fft.

I guess it doesn't really matter what I run it at though since I'm getting a 45nm dualy pretty soon, which I hope can do 4.2Ghz.

XDanger
02-01-08, 01:44 PM
9650 @ 4.4Ghz @ 1.56v 4.2Ghz @ 1.50 24/7 but on a very good water setup.


!OVERCLOCKING!
Teh Sport Of Nerds(nana2)

Nice :D .