Voltage Max vs Cooling on 6400c4's

[SriGyre]

I have 2 gigs of XMS2 PC6400c4 memory, I would like to know if I can run them past 2.1V for OC'ing purposes. I would like to tighten timings to 4-4-4-12 while taking the memory over 1100mhz. This may only be a pipe dream, however, I would like to make the attempt, and would like some feedback. I am water cooling the sticks along with my chipset/mosfets, CPU, and GPU's. My MB will alow me to put 2.5 Volts to the memory. If I have them watercooled can I get away with this much juice?

[Wired]

warrantied up to, but not over, 2.1 volts

[DerekT]

I have 2 gigs of XMS2 PC6400c4 memory, I would like to know if I can run them past 2.1V for OC'ing purposes. I would like to tighten timings to 4-4-4-12 while taking the memory over 1100mhz. This may only be a pipe dream, however, I would like to make the attempt, and would like some feedback. I am water cooling the sticks along with my chipset/mosfets, CPU, and GPU's. My MB will alow me to put 2.5 Volts to the memory. If I have them watercooled can I get away with this much juice?

 

I have these same modules. I use them with a Gigabyte 965P-DS3 and Core2 E6400 @ 3.2GHz. I have tested these modules up to 1000MHz at 2.1V and 5-5-5-15. Remember that electrostatic discharge is a phenomenon not contingent upon heat output and can injure CMOS regardless of temperatures when the voltage is high enough.

 

There is a condition called "Threshold Voltage" which is the amount of power necessary to turn on and switch the transistors gates. There is a formula for the switching power:

 

Power ~ ½ CV ² Af

 

This formula states that dissipated power is linear with the effective capacitance, activity and frequency. Power need increases quadratically with the voltage and output of the voltage will create a transference effect which can (both short term and long term) damage the CMOS.

 

Thus when you overclock any CMOS, you are increasing the necessary voltage by a multiple of four and the voltage effects are not linear but multiplicative, quadratic and exponential. These effects can quite often lead to degradation of the CMOS.

 

This does not mean that you will damage your chips in the shorter term. All CMOS is imbued with a half life, and as such is not a normal method of gauging life.

 

Silicon Substrates as well as the CPU/DRAM chips do not have a life such as humans and animals etc. have. In their physical nature they are endowed with a half life, so you can consider their span as radioactive decay for purposes of determining possible length of minimal error use.

 

For example:

 

Given:

 

10,000 New CMOS's from the same fabrication silicon wafer. Their half life is the time that would pass before 5,000 of them fail. So, based on how long people have working CMOS's of the same fabrication silicon wafer, the half life of your CMOS could easily be 10 or more years (non overclocked).

 

This is why the warranty is given to 2.1V. It does not mean that you will damage the CMOS in the shorter term. It does mean that you will take all responsibility for the possibility of damage and you can be certain that the dram can be tested for high overvolts and a void of the warranty will be forthcoming.

 

It's all up to you.