I thought I'd assemble it first then install the OS. Actually doesn't matter, RAM is not that difficult to swap or change.

 

Thanks,

Dirk.

Here are my idle temps after delidding and using Antec Formula 7 Diamond paste. Core 2 must be on vacation. :huh:

 

Cheers

jja

Here are my idle temps after delidding and using Antec Formula 7 Diamond paste. Core 2 must be on vacation. :huh:

 

Cheers

jja

 

Either that or the other 3 cores aren't getting enough cooling lol :-D

 

Sent from my Galaxy Nexus using Tapatalk 2

 

 

Yea I figure that it is sitting on the sweet spot of the H100 cooling block. All cores ramp up to similar temps under load though (<70c)

 

Cheers

jja

To those with Maximus motherboards, please, where is Turbo Voltage Setting in BIOS?

 

I refer to this explanation from overclock.net (based on asrock mobo):

 

Intel Turbo Boost Tech: Enabled

~This lets us use Additional Turbo Voltage.

 

Additional Turbo Voltage: +0.004v

~This setting will be changed later.

~This is just like the Offset but works ONLY when the CPU is not in idle state.

~The Offset works ALL the time, even at idle. This setting will allow you to keep a low Offset, and low idle voltage, while still getting the Vcore boost needed for full speed.

==================================================

 

 

On a side note: I'm still working on my new build (i7 3770k + Corsair H100i push-pull + Maximus V Extreme ) and so far it seems to be another successful de-lidding story. In AIDA64 @4.8Ghz (Offset mode, C3 and C6 off) temps never exceed 67C/ vs.22C ambient) even though my chip is not that good, it requires 1.416V vcore (VID for that clock is 1.346v in Core Temp) to run stable in offset mode (+0.065) and 1.400V in fixed mode under 100% AIDA64 load. I'm trying to push it higher to 4.9 or even 5.0 in offset mode, but but can't find the right settings. Thus Turbo Voltage setting above seems for a reason in my case.

A couple of 'theoretical' questions:

 

1) How did you apply Liquid metal to your chip? I just spread it with a cotton swab both on the chip and IHS. On overclock.net it is recommended to leave just a tiny drop as when you start spreading it the TIM will have internal air bubbles (or smth. like that).

2) Did you disable Write-caching policy on your OS SSD when doing and testing your o/c? Could be a very good idea.

3) Did you get occasional 'COM Surrogate has stopped working' error when testing your o/c? I understand that could be due to anything ranging from missing codecs to errors on you os ssd and hw instability.

A couple of 'theoretical' questions:

 

1) How did you apply Liquid metal to your chip? I just spread it with a cotton swab both on the chip and IHS. On overclock.net it is recommended to leave just a tiny drop as when you start spreading it the TIM will have internal air bubbles (or smth. like that).

2) Did you disable Write-caching policy on your OS SSD when doing and testing your o/c? Could be a very good idea.

3) Did you get occasional 'COM Surrogate has stopped working' error when testing your o/c? I understand that could be due to anything ranging from missing codecs to errors on you os ssd and hw instability.

 

1. Spread it with the cotton swab or brush as instructed by Coollaboratory on the die ONLY. You can apply it to the IHS as well, but only on top, where the IHS meets the heat block. No need to apply it to the heat block though.

2. I would highly advise you do NOT apply any tweaks until you are sure your system is fully stable. As a matter of fact, make sure you are able to boot into Windows and perform some STOCK stress testing to make sure none of your components have issues. Once you know your system is up to par STOCK, then you can start OC'ing. Once your OC is done and stable, then you can start messing around with tweaks ONE at a time. Saves you a ton of headaches when things start to fail since you will be able to know what the culprit was.

3. If you were not getting these errors when stock, then yes, it might have to do with the OC. This is why step 2 is SOOOOO important.

 

Good luck, Dirk!

I don't believe enough pressure is created to just put a dot of liquid metal on the CPU die and pop the lid back on. I do not recommend this method at all. You need to ensure a good contact patch.

1) COM Surrogate error was due to something else I did at the very beginning (messed with Folder View)

 

@BimmerCop

2) I understand Disable Write-caching policy is a good advise BEFORE any overclocking attempts as "equipment failure might result in data loss or corruption" when this poicy is On (by Default). And equipment failures may happen plenty while overclocking thus better to disable it before messing with o/c.

 

@TechGuy:

3) Do we have Turbo Voltage on Maximus motherboards?

 

Thanks! :smile:

 

Dirk.

I don't believe enough pressure is created to just put a dot of liquid metal on the CPU die and pop the lid back on. I do not recommend this method at all. You need to ensure a good contact patch.

 

For what it's worth... I place a rice-sized grain (OK, a big rice-sized grain) of Liquid Metal Ultra on the die, replaced the lid for a test fit, and didn't see great contact between the two. So, using the brush, I picked up a little of the Ultra from the die and painted it on the underside of the lid. Didn't add any, just redistributed some to the inside of the lid. Also treated the top of the lid, but not the waterblock. Am getting very good results - temps around 60 in FSX, and no more than 70 in AIDA64.

 

@Dirk98 - sorry, I'm not on a Maximus, I'm on a P8Z77-V Pro, so can't help with the turbo voltage. But there are plenty of Maximus users here so with luck, somebody will chime in.

For what it's worth... I place a rice-sized grain (OK, a big rice-sized grain) of Liquid Metal Ultra on the die, replaced the lid for a test fit, and didn't see great contact between the two. So, using the brush, I picked up a little of the Ultra from the die and painted it on the underside of the lid. Didn't add any, just redistributed some to the inside of the lid. Also treated the top of the lid, but not the waterblock. Am getting very good results - temps around 60 in FSX, and no more than 70 in AIDA64.

 

Alan, what is your cooler?

 

Thanks,

Dirk.

Dirk -

 

It's an XSPC custom watercooling setup. I started with this kit, then read a little more and decided to upgrade to their D5 Vario pump and enough extra tubing to get my graphics card into the loop. Might be overkill - Hasse Westman, who knows more about cooling than any of us, has said that the closed-loop systems like the Corsair H100 are the best value, and you don't gain much from a custom loop. But I enjoyed the process of figuring it out and putting it together, which is something, and the results are pretty decent. From what everybody here has said, it looks like I was pretty lucky with the chip, too - it's a much-better-than-average performer.

 

Hope that helps.

Have you de-lidding fans ever considered an oven?

 

Not as daft as it seems. TJMax may be the limit for a chip that's powered up, with current passing through, but from what I understand, the temp limits are far higher for an un-powered chip. I gather during mmanufacture the chips get very toasty indeed.

 

Seems to me, that warming the CPU up in your kitchen oven, would soften the thermal adhesive and render the delidding process far easier.

 

Don't tell me off if you screw up your CPU though. :smile: It won't be my fault. I will deny all responsibility. Your risk not mine.

Have you de-lidding fans ever considered an oven?

 

I like it! Sounds like I missed a step!

 

Seriously... though I'm getting good results with my de-lidded CPU, I'm not sure de-lidding is necessary, and I'm sure it's not for everybody. Part of the reason I did it was for the sheer craft-oriented, transgressive fun of taking a razor blade to a piece of expensive electronic equipment. Back when I was first contemplating de-lidding, I posted something about my first computer upgrade - back in the late 1980's, I installed the Dove Memory Upgrade into my then-top-of-the-line "Fat" Macintosh (fat meaning that it had 512k of RAM, rather than the base 128k). The Dove upgrade took it to a full gigabyte. To install it, you had to use a spreader to pry the back of the case off your Mac, which exposed the naked back of the CRT, which, being a vacuum tube, if you punctured it accidentally, it would explode in your face. Then you had to shave down the mounting clips on the memory modules with a razor. I remember sitting there at my desk with my safety goggles on, sweating and waiting for the explosion... which of course never came. But it was a real thrill when the system booted up with 1024k RAM on board. I've never quite recovered from that and I've been looking to (physically) hack systems ever since. The de-lidding process gave me that kind of thrill - probably way out of proportion to its impact on my FSX performance.

 

But then, without the hobbies-within-the-hobby, where would we be?

Alan, I agree with you entirely. I decided for me, I couldn't justify de-lidding. But yes, I agree, for the "fun" and sense of achievement in terms of bypassing Intel's TIM lunacy, I can certainly see the appeal. :smile: