OVERCLOCK PHENOM II X4 965

[trents]

kilyan, spreading the paste like you did with your finger, even with it wrapped in cellophane has a way of introducing air bubbles which act as insulators.

The best way for a cooler with a flat plate bottom like yours is to squeeze a pea-sized blob in the center of the CPU face and then just lower the cooler on and clamp it down. Let the clamping pressure spread the paste. If the clamping mechanism is a bolt through type then tighten each screw a little at a time with a back and forth "X" pattern to create even pressure. But first, clean off the old paste well with paper coffee filters as they don't leave insulating fibers like soft cloth or tissues due. Start with dry filters to remove the excess and finish up with one moistened with rubbing alcohol. Clean both the CPU face and the heat sink bottom well.

Attention to little details like that can make a difference sometimes.

 

[kilyan]

Most of us when using a perfectly flat (No Exposed Heat Pipes), use a rice grain sized dab of AS5 in the center of the processor. A Lentil sized dab of AS5 in the center of the processor is also used.

Most of us do not use any other method. I use a dab in the center, only, myself.

so the rice grain(nugget) sized paste spread by the pressure between cpu and cooler is the right way if i did catch right?...

 

[kilyan]

Most of us when using a perfectly flat (No Exposed Heat Pipes), use a rice grain sized dab of AS5 in the center of the processor. A Lentil sized dab of AS5 in the center of the processor is also used.

Most of us do not use any other method. I use a dab in the center, only, myself.

kilyan, spreading the paste like you did with your finger, even with it wrapped in cellophane has a way of introducing air bubbles which act as insulators.

The best way for a cooler with a flat plate bottom like yours is to squeeze a pea-sized blob in the center of the CPU face and then just lower the cooler on and clamp it down. Let the clamping pressure spread the paste. If the clamping mechanism is a bolt through type then tighten each screw a little at a time with a back and forth "X" pattern to create even pressure. But first, clean off the old paste well with paper coffee filters as they don't leave insulating fibers like soft cloth or tissues due. Start with dry filters to remove the excess and finish up with one moistened with rubbing alcohol. Clean both the CPU face and the heat sink bottom well.

Attention to little details like that can make a difference sometimes.

i don't think have that coffee paper filters,i have cotton-fioc, the thing you use to clean ears

 

[trents]

No, don't use any cotton stuff; very soft and loose and will leave behind fibers. Newspaper would be better than that. Or gift wrapping tissue paper would work. Anything with a smooth, hard sheen would do.

 

[kilyan]

No, don't use any cotton stuff; very soft and loose and will leave behind fibers. Newspaper would be better than that. Or gift wrapping tissue paper would work. Anything with a smooth, hard sheen would do.

toilet paper maybe?
anyway my first tests for 3.8 ghz:

 

[kilyan]

prime 95 result

 

[trents]

No, toilet paper is too soft. Don't you have any newspaper around? Or a paper bag would probably work well.

Anyway, temps are looking decent at present configuration. Now see if you can pass a 2 hr. Prime95 blend test with that 1.392 vcore. It's still pretty low and you may need to bump it up to get it truly stable.

 

[trents]

kilyan, when are you going to create a "Sig"?

 

[kilyan]

No, toilet paper is too soft. Don't you have any newspaper around? Or a paper bag would probably work well.

Anyway, temps are looking decent at present configuration. Now see if you can pass a 2 hr. Prime95 blend test with that 1.392 vcore. It's still pretty low and you may need to bump it up to get it truly stable.

yes paper bag i have! ok will try to test it 2 hours now with prime 95

 

[kilyan]

kilyan, when are you going to create a "Sig"?

where do i have to go to create one?
edit:nvm found now ill post my specs as Sig

 

[kilyan]

here it is

 

[trents]

Thanks.

 

[kilyan]

yes paper bag i have! ok will try to test it 2 hours now with prime 95

after 1 hour still stable and 53 degrees max

 

[kilyan]

after 2 hours test passed succesfully,now what?

 

[trents]

Well, you can go for 3.9 ghz. It's possible you may have temp room for one more small increase in vcore.

 

[kilyan]

Well, you can go for 3.9 ghz. It's possible you may have temp room for one more small increase in vcore.

how? increase multiplier to 19.5 and vcore to 1.4?
intelburntest gave error.increase voltage one step more?

 

[caddi daddi]

coming along very nicely kilyan.

 

[trents]

how? increase multiplier to 19.5 and vcore to 1.4?

Give it one more small bump if temps will tolerate it.


intelburntest gave error.increase voltage one step more?

Don't forget to test it with Prime95 blend as it does a better job of stressing the memory and integrated memory controller component.

 

[kilyan]

really?thought intelburn test was heavier cause it tests also all ram,while prime 95 tests only a small part of ram.
Anyway at 1.4125 3.9 ghz passed through 10 cicles intelburntest succesfully,ill try prime 95 now

 

[kilyan]

really?thought intelburn test was heavier cause it tests also all ram,while prime 95 tests only a small part of ram.
Anyway at 1.4125 3.9 ghz passed through 10 cicles intelburntest succesfully,ill try prime 95 now

first half an hour 56 degrees max
success after 1 hour prime 95: