The TRUTH about Prescott Overclocking...

[Jeff7181]

Originally posted by: rabidoverclocker
My 2.4A runs at 44C idle and 52C-56C under full load at 1.7 volts (which is always around 1.55-1.6) on my AI7. All of you saying 1.7 vcore is too much are most likely amateurs...

Im using a custom watercooling setup w/ a DTEK Whitewater. There is no advantage upping the voltage past 1.7, it does not yield any overclocking benefit.

I can run this chip @ 3.8 (1.8 vcore) but i keep it at 3.6 to keep the temps/workload on the mofsets down.

I've used my AI7 problem free for about a month, most of the time running prime95 torture and it gives me no problems.

I went and got a new MSI 865PE NEO2-PLS and set it to 18X200 (3.6ghz) with 1.6 vcore and the motherboard failed (burnt mofsets) just like the other one in less than 30 minutes under Prime95 Torture Test.

This AI7 can handle fine what TWO brand new MSI boards could not...go figure...

If you guys can't see lack of quality on MSI boards "supporting" the Prescott then that's your deal. You can physically see they skimped on the power regulation just looking at the board compared to the AI7. The solder points on the MSI board show they used only HALF of the capacitors, mofsets, etc. that the board was originally designed for in order to cut cost...and ultimately ruinined the quality of the board for serious overclocking...

You won't see ABIT, ASUS, etc boards burning up like this...

Well there goes your credibility...

"increasing voltage past 1.7 doesn't yield any gains" ... "if I increase the voltage another .1 volts I get 200 MHz more"

 

[jhurst]

I wouldn't worry about your temps if you are running a watercooled system. Your actual temps are going to be cooler than Hans temps, even though he is running a Northwood.

Abits boards are holding together better than MSI b/c they use better quality parts in them (i.e. Rubycon capacitors). That isn't to say that MSI is crap, they just aren't building good boards for the extreme overclocker. I would go with nothing beside Abit or Asus for an OC'ed system (oh yea, and maybe a DFI).

 

[rabidoverclocker]

Originally posted by: Jeff7181

Originally posted by: rabidoverclocker
My 2.4A runs at 44C idle and 52C-56C under full load at 1.7 volts (which is always around 1.55-1.6) on my AI7. All of you saying 1.7 vcore is too much are most likely amateurs...

Im using a custom watercooling setup w/ a DTEK Whitewater. There is no advantage upping the voltage past 1.7, it does not yield any overclocking benefit.

I can run this chip @ 3.8 (1.8 vcore) but i keep it at 3.6 to keep the temps/workload on the mofsets down.

I've used my AI7 problem free for about a month, most of the time running prime95 torture and it gives me no problems.

I went and got a new MSI 865PE NEO2-PLS and set it to 18X200 (3.6ghz) with 1.6 vcore and the motherboard failed (burnt mofsets) just like the other one in less than 30 minutes under Prime95 Torture Test.

This AI7 can handle fine what TWO brand new MSI boards could not...go figure...

If you guys can't see lack of quality on MSI boards "supporting" the Prescott then that's your deal. You can physically see they skimped on the power regulation just looking at the board compared to the AI7. The solder points on the MSI board show they used only HALF of the capacitors, mofsets, etc. that the board was originally designed for in order to cut cost...and ultimately ruinined the quality of the board for serious overclocking...

You won't see ABIT, ASUS, etc boards burning up like this...

Well there goes your credibility...

"increasing voltage past 1.7 doesn't yield any gains" ... "if I increase the voltage another .1 volts I get 200 MHz more"

3.8 isnt completely stable, prime95 will error out after about 8-10 hours on average running at that speed/voltage. i said that more to show that the cpu can be run at 1.8 volts and not "die"

the highest completely stable i have gotten with this chip is 3.7 using 1.6 volts...sorry for the confusion

 

[Jeff7181]

Originally posted by: rabidoverclocker

Originally posted by: Jeff7181

Originally posted by: rabidoverclocker
My 2.4A runs at 44C idle and 52C-56C under full load at 1.7 volts (which is always around 1.55-1.6) on my AI7. All of you saying 1.7 vcore is too much are most likely amateurs...

Im using a custom watercooling setup w/ a DTEK Whitewater. There is no advantage upping the voltage past 1.7, it does not yield any overclocking benefit.

I can run this chip @ 3.8 (1.8 vcore) but i keep it at 3.6 to keep the temps/workload on the mofsets down.

I've used my AI7 problem free for about a month, most of the time running prime95 torture and it gives me no problems.

I went and got a new MSI 865PE NEO2-PLS and set it to 18X200 (3.6ghz) with 1.6 vcore and the motherboard failed (burnt mofsets) just like the other one in less than 30 minutes under Prime95 Torture Test.

This AI7 can handle fine what TWO brand new MSI boards could not...go figure...

If you guys can't see lack of quality on MSI boards "supporting" the Prescott then that's your deal. You can physically see they skimped on the power regulation just looking at the board compared to the AI7. The solder points on the MSI board show they used only HALF of the capacitors, mofsets, etc. that the board was originally designed for in order to cut cost...and ultimately ruinined the quality of the board for serious overclocking...

You won't see ABIT, ASUS, etc boards burning up like this...

Well there goes your credibility...

"increasing voltage past 1.7 doesn't yield any gains" ... "if I increase the voltage another .1 volts I get 200 MHz more"

3.8 isnt completely stable, prime95 will error out after about 8-10 hours on average running at that speed/voltage. i said that more to show that the cpu can be run at 1.8 volts and not "die"

the highest completely stable i have gotten with this chip is 3.7 using 1.6 volts...sorry for the confusion

There's people who have run Bartons a T-Bred B's at 2.1 volts for a while too... that doesn't mean they won't die if you leave them at that voltage.

 

[rabidoverclocker]

Originally posted by: Jeff7181

Originally posted by: rabidoverclocker

Originally posted by: Jeff7181

Originally posted by: rabidoverclocker
My 2.4A runs at 44C idle and 52C-56C under full load at 1.7 volts (which is always around 1.55-1.6) on my AI7. All of you saying 1.7 vcore is too much are most likely amateurs...

Im using a custom watercooling setup w/ a DTEK Whitewater. There is no advantage upping the voltage past 1.7, it does not yield any overclocking benefit.

I can run this chip @ 3.8 (1.8 vcore) but i keep it at 3.6 to keep the temps/workload on the mofsets down.

I've used my AI7 problem free for about a month, most of the time running prime95 torture and it gives me no problems.

I went and got a new MSI 865PE NEO2-PLS and set it to 18X200 (3.6ghz) with 1.6 vcore and the motherboard failed (burnt mofsets) just like the other one in less than 30 minutes under Prime95 Torture Test.

This AI7 can handle fine what TWO brand new MSI boards could not...go figure...

If you guys can't see lack of quality on MSI boards "supporting" the Prescott then that's your deal. You can physically see they skimped on the power regulation just looking at the board compared to the AI7. The solder points on the MSI board show they used only HALF of the capacitors, mofsets, etc. that the board was originally designed for in order to cut cost...and ultimately ruinined the quality of the board for serious overclocking...

You won't see ABIT, ASUS, etc boards burning up like this...

Well there goes your credibility...

"increasing voltage past 1.7 doesn't yield any gains" ... "if I increase the voltage another .1 volts I get 200 MHz more"

3.8 isnt completely stable, prime95 will error out after about 8-10 hours on average running at that speed/voltage. i said that more to show that the cpu can be run at 1.8 volts and not "die"

the highest completely stable i have gotten with this chip is 3.7 using 1.6 volts...sorry for the confusion

There's people who have run Bartons a T-Bred B's at 2.1 volts for a while too... that doesn't mean they won't die if you leave them at that voltage.

A 2.4 Prescott only costs around $125

A Barton or T-Bred is even less...

There is no reward without risk...

MSI boards are the suck right now for overclocking, not on features, but on quality