DDR3 RAM

I see! That's why they are more expensive than other 2500s. If I wait for a month or two, do you think they will still be available?

Yes they'll still be available.

Intel's next lot of new processors (Ivy Bridge) seem to be scheduled for a 2012 release.
 
Last edited:
So what actions should I take, I just put the new H60 in. It's quite painful to do it again...

Run Prime95 on all 4 cores and note the difference between the core temperatures at load.

If it's a few degrees difference then that's normal.

If the difference is more than a few degrees it really does suggest the cooler isn't sitting properly on the CPU.

Perhaps you tightened one corner or one side more than the other.
 
Surveyor
I ran the test, horrific!
Now @ 3.6GHz, both core1 and 2 reported 82C
I stopped immediately.
This is now thei dle temperature

I'll go tighten the screws now....

 
You probably need to do a bit more than just tightening the screws then.

It looks like you'll have to remove the cooler block from the CPU, re-apply the thermal paste and then re-fit it very carefully.
 
You probably need to do a bit more than just tightening the screws then.

It looks like you'll have to remove the cooler block from the CPU, re-apply the thermal paste and then re-fit it very carefully.

This problem exists before the H60, I was using the Silent Knight. Core1>Core2>core3>core4

Just now I have ran Prime95 again. Core 1 reached 98C where as Core 4 was 68C ridiculous!! A difference of 30 degrees.

I am tired I need to get up at 7am. I guess I will have to do it tomorrow after work.

But if I reapply thremal grease I will lose the original grase from Corsair, I heard the default grease is very good grease.

I have Thermaltake TG-1 standing by, is it a good brand?
 
On second thought Surveyor,
Could that be the grease covering Core 1 and Core 2 are too thick making hard to conduct heat? Where as Core 3 and Core 4 are just about right thickness?

There is a map about the orientation of all 4 cores...do yo have the link?
 
Last edited:
Just a quick search gave me this
http://www.xtremesystems.org/forums...roper-Quadcore-water-block-mounting-technique

Surveyor,

Something I am not sure. Is that the thicker the grease on a core, the hotter it will get?

Is there a hard and fast rule for distributing grease over the CPU ? Also If I am going to repaint, should I put the grease on the CPU of on the heat sink?

The thermal paste should spead out under the pressure from the cooler unless you have a cpu with an oddly shaped heat spreader.

The thermal paste is normally applied to the CPU. Not too much though.

Here's one suggestion on how to do it with a thin line across the 2 pairs pf cores.

 
Last edited:
Surveyor

I opened up the sink and found core 3 and 4 are not covered at all!!!!!! That's the default thermal paste on the H60

I reapplied the paste, it's thermaltake.

They are now closer in temperature.

But temperature seems to be higher or they seem alright to you...check for me I am tired....
Thanks!!!

 
If that's still at 3.6GHz then the temperatures are OK(ish) but you don't want to be any higher and a few degrees lower would be better.

Ideally Cores #0 and #1 should nearer to the temperatures of Cores #2 and #3.

It looks like your cooler and CPU just don't want to sit entirely happily with each other.
 
If that's still at 3.6GHz then the temperatures are OK(ish) but you don't want to be any higher and a few degrees lower would be better.

Ideally Cores #0 and #1 should nearer to the temperatures of Cores #2 and #3.

It looks like your cooler and CPU just don't want to sit entirely happily with each other.

A new day has begun, good morning Surveyor.

You mean the overclocked temperature is okay? That's touching onto 80C.....

Guess what happened before the rework of the cooler: I opened it, half of the CPU surface (assuming core 1 & core 2 area from your picture) was covered by the default paste, HOWEVER, you might not believe this, the other half (i.e. core 3 & core 4) was not! The default paste was on the cooler's copper surface instead. Having said that, as we can read from the HWMonitor, the temperatures of both core 3 and core 4 are lower than core 1 and core 2.

After the rework, now that core 4 is still the lowest. What I did was I made the paste from core 1 through to core 4 in a line, stirred around until almost the entire CPU surface is covered by the paste.

This morning, I fired up the machine again. Core 1 and Core 2 were both at 46C and core 4 was 39C.

You know what I wish? I wish to bring core 1 through 3 down to 39C. If core 4 can stand at 39C at idle state, why not the other three, it's the same paste afterall! What have I done wrongly?
 
Good morning.

It may be a new day for you but it's after 2am here and I'm off to bed!

Just having a last look at the forum.

80 degrees is a far as you want to go and, as I said, it would be better if it was a bit lower.

If the paste hasn't spread evenly over the CPU then the cooler isn't making proper contact.

Either:

1) You're not fitting it properly

2) The CPU heatspreader is an odd shape

3) The base of the cooler isn't flat

If it's options 2 or 3 there isn't a lot you can do.

If you want to try re-fitting the cooler (again) that's up to you.

Perhaps try turning the cooling block through 90 degrees to see if you get a better fit.

If you have an old cooler you could try fitting that. If the thermal paste spreads out evenly then that indicates your problem is options 1 or 3.

If you want to refit the H60 have a look at this video, it may assist.

I really do have to go to bed now.

I'll check back later.

Night, night.
 
Last edited:
Final thing before I go.

Here's another diagram of the core numbering.

It might help you figure out which are the hotter cores.

It really is time for bed.

Good night.

 
The best results I have got! And a bad news!!

Hi Surveyor,

Hope you have a great day ahead!

For the last 3 hours, after leaving office I have tried out 6 re-fitting the H60, the TG-1 grease almost used up I guess.

You mentioned before may be the thermal distribution inside the chip itself is sort of polarised, no matter how much more or how much less I use, the TG-1 helps, I come to that later; but the core 0 and core 3 polarization is like a constant.

I applied more grease to my first few attempts, and finally managed to strike a balance between the thickness of the coating and the holding idle temperature. It is slightly higher with thicker grease, but the temperature reaches a peak of 72C running Prime95 and that's the highest it goes. With thinner coating, the initial temperature is lower only slightly though, but when it comes to Prime95, it goes as high as 78 to 80C!

After 7 attempts, here is the result I have got




Also, i think we are the luckier folks in that we got a pretty good overclockable chip, but it never goes beyond 3.6GHz. At 3.7GHz, it boots but one of the cores failed the Prime95 stress test. At 3.8GHz, it boots but 3 cores failed the Prime95 test I bet if i continued, the last core would have failed also.

So that's all, the final verdict is 3.6GHz, that's all I have got. Stable for 20 minute so far and the highest is like 72C, is that temperature acceptable?
 
Bad news is I was in a hurry to do the refit, and I broke the front hatch of the left panel of my beloved 600T!!!!! Well the panel still fits to close but opening it require some manual adjustment because the front hatch is no longer operable!

Oh.....my HK$ 1,280!!!
 
Back
Top Bottom