Official ICD7 Thread

GO FOR IT!!!

 

No No No No........you cannot lap a core!!!!! Dont even joke about stuff like that, because there are people out there who do not know better.

Whatever you do, do not lap an exposed core, be it the CPU or GPU.

 

Eight more samples hit the mail today US and international just to let you people know it's on it's way

 

Biggest bang for temp improvement I have seen other than improving contact by lapping is increasing pressure Out of 10 forum giveaway's the resulting troubleshooting guide boiled down to Contact or Pressure
as the primary cause people did not see a result. Works for all pastes not just IC Diamond.

The Intel stock sinks run from my memory 43 +/- 10lbs at below 50lbs the thermals comparison tend to get a little homogenized. May be worth a shim if you are sitting at 30 PSI you will gain a few degrees.

 

With an IHS thats true, but with an exposed core like those found on notebooks, be very careful....you can crack the silicon and ruin your system.

 

A Summary of results to date- It helps if you list before and after results and do not forget ambient temp as it is sometimes hard to decipher and provides a little clearer picture

 

haha my temps kinda wrecked the average there :|

will have GPU temps up momentarily

 

-4.479 C without it

 

Exactly what happened with my old AMD Duron 1600Mhz Chiped a corner
And ofc- YOU SHOULD NEVER LAP A CPU OR GPU CORE So everyone sees it.

 

Can't wait to see how this stuff will perform on my desktop CPU's in the C90.

 

u should see a noticeable decrease in temps if applied correctly...........

 

Woo glad to hear the samples are on there way, and so is my new lappy =D man this is a good kick off to a great summer!

 

Im glad I got to participate in the free samples, I will put it on my CPU & both GPU's on my W90 when its back from RMA.

Didnt somebody say this stuff is thick enough to also replace the thermal pads on the GPU's? I dont know if I want to try that or not so I better find some good thermal pads to put on there too.

Anybody have recommendations for thermal pads that are high quality to pair with the ICD7?

The W90 will take great advantage of the better compound as it has a robust cooling system, if your cooling is not up to the task to wick away the heat fast enough, increasing the efficiency of transfer from the components to your heatsink really wont do any good.

 

The ICD 7...

 

So it does work... hmm but considering people have had issue getting it to apply well on just there core with almost no gap, trying to cover a big gap seems risky, and i doubt my free sample will have enough to cover 2gpu cores, a cpu and like 12 ram chips.

 

I'd order a tube of the ICD-24, which is their bigger tube. If you want to go with thermal replacement pads, Shin Etsu makes some really good ones.

If all else fails, you could always use the copper shim mod that many people used before learning about ICD7.

 

I sent him a tube of the ICD24 Small test group - big apps-big tube

 

Oh sweet, well then I guess I owe it to you to try to cover the ram as well.

Im trying to think of a good way to accurately measure the real distance between the heatsink and the ram chips before applying it, I guess just a dummy test applied to one chip and nothing else and remove the heat sink to see how much it was compressed. Good times!

 

OMFG!!!!!! DUUUUUUDE! this thing ROX. Damn 9C less, and I surely made the highest gpu overclock I could think of... .

I just got the paste today and I couldnt wait any longer to apply it .

Look at these temps after playing COD W@W for 40 mins... Note I used my gateway that is in my signature!.

Mild overclock (ati HD3650) with AC5:


Mad overclock (ati HD3650) with ICD7:


Laptop at idle AC5:


laptop at idle ICD7:




This thermal paste is the bomb, the new law , I cant believe it I got like 10c less in my gpu!!.

The ONLY downside is that this paste is VERY thick and is really hard to apply but you can put it the tube on warm water first...

Thanks again for the paste, Im gonna post more results later!

 

Wow, those results are excellent. The temp jump on your GPU is ridiculous. Glad it worked for you . I can't wait to try it out now.

 

The temp improvements you guys are getting great. I am thinking I might pull a couple of mine apart and check the coverage pattern. I'm thinking it might be a little thin. I did see temp reductions, but I might get better yet. Plus, I can apply it to the GPU ram and see if that makes a difference.
Note that all my tests where run with the fans forced to high, for both idle and stressed temps.

 

Pfttt... and this is getting better and better... I saw that I didnt apply the thermal paste in the correct way, so I decided to reapply it today.

I followed the steps carefully, those steps that are at the ICD webpage, I also considered to apply it on the northbridge since there was a thermal pad there, the only parts in my heatsink without ICD are the gpu memory heatsinks, since I felt that they couldnt make full contact to the metal. And well the results were more than stunning... ICD has my FULL respect...

Take a look at this, I ran 3dmark06 like 15minutes, COD W@W 45 minutes, this with a mad overclock(same as the last time)... These are the temps at full load:


Before (wrong application):




And you will ask me: Whats the difference from my last paste application?
Northbridge: -4C
Processor: -4C
GPU: -3c

Wow thanks again for the thermal paste, it seems to like my laptop . Ill do more test with other computers later.

 

Hmm, I could really use some of this thermal compound for my EliteBook laptop -- it used to be damn quiet and cool, but once I had the defective GPU replaced, it's become far warmer under load; it seems HP used less-effective thermal tape or grease.
So, I just send my mailing address to the e-mail address in the first post? I'd be glad to post test results for the thing, for apps such as Crysis, perhaps.
Interestingly enough, the EliteBook has several ACPI thermal zones, in addition to the CPU and GPU die sensors: BATZ, CP2Z, CPUZ, DTSZ, GFXZ, LOCZ.

 

Sorry if I missed it somewhere. I'm currently using AS5 and want to try IC7, who should I talk to and how can I obtain a sample?

 

read the 1st post '

 

Whoops, was tired. So fail on me Thanks, sentence.

 

I just posted instructions to get to the GPU here.

[thread]382408[/thread]

 

cpu & gpu are the same thing, just different cores, directions should be the same. Since this is non conductive its better for gpu uses than say AS5 and if it managed to ooze out some from compression it wont hurt anything.

My biggest take on reading the official install directions is its better to use too much than too little.

 

Exactly, here aplies the saying: "too much is better" And for 1 mm gap between the core and heatsink, the IC7 works just fine. I had that gap with my laptop NB and IC7 worked wonders

 

MX-2 cutted my temps noticeably, ordered one ICD7 over eBay 4 days ago, now I can barely wait to see the improvements I'm going to apply it everywhere (CPU, North Bridge, GPU, vRAM), I hope that it would have enough paste in there though. My GPU is currently copper moded, so I'll test it with and without the copper sheet, if there is enough paste

 

I just modded a laptop from a friend it was the xps m1530, I dont have screenshots yet but I can tell that temps lowered at least 5C .

 

yes this magic stuff works wonders those that say that the temps have gone up instead of down are applying it incorrectly......or at least thats what it seems like.....

 

Yeah, they should put the tube in HOT water for a few minutes first , then when it is more like a liquid you apply it just squeeze the size of a rice on the processor, very important to follow the instructions at the ICD7 webpage.

 

yeah i knoe i applied some on my friends laptop as well and the temps went up instead of down so i had to reapply and the temps went down 4c i put to much in the beginning learned my lesson

 

hmm, I sent an e-mail on Saturday or Sunday... but haven't gotten a response yet.

 

They will send it out to you without sending you a reply. Mine arrived today and I only sent a PM to ICDiamond on the 13th

 

I got mine yesterday, my wife opened it on my behalf and was like "what is this???"

Now I just have to wait for my laptop....

 

Man, would that be cruel.....your wife sees some package with "Diamonds" on it. Might be a good joke for a wife someday (not mine, not brave enough). Get her a big tube of ICD24, tie a loop of string on it, and tell her you bought her a 24K diamond necklace.

 

Just tested the ICD 7 on my old E6600 in the C90 and can report decent results. Because of the way in which the fans work in the C90 one cannot report temp changes but rather the effectiveness of the paste is reflected in the behaviour of the fans.
To cut a long story short, before using old paste the temps would shoot quickly to over 70C before the fans kicked into high gear and pulled them down to the 70C permanently. With iCD 7 the temps climbed much more slowly to 70C eventually reaching 71C at which point the fans pulled the temp down to 70C BUT it never had to kick into highest speed to do so when they had to before.

I conclude a positive result therefore and probably in effect at least a drop of 5C if only I could control the fan speed for a fair test.
Idle temps have dropped 2C from max of 54C to 52C

Will put some on the GPU when I have time. Should be even more positive!

 

haha perfect mother's day prank. I received my tube sample a few days ago. I'll put it on the CPU once I have my computer backed up. I'm not sure if I'm daring enough to tear my notebook apart to get to th GPU.

 

She did tell me "Man what a waist of diamonds" lol.

 

So my 5796 just arrived at my gf's, as well as the ICD7. She thought I had bought drugs because of the syringe...im just like...who do you take me for? XD Anyways will get the temps up ASAP.

 

LOL you should have gone along with that! haha

 

She prolly asked because he had the same goofy look on his face as Bob from Enzyte:

 

Well my friend Elias just sent me this for you guys .

Cooling Results:

Technical Data:

 Dell XPS 1530
 Bios A12
 Intel Core 2 Duo CPU T5750 - 2.00 GHz
 RAM 3.00 GB DDR2
 32 bits OS
 Nvidia Gforce 8600 GT 256 MB
 250 GB for HDD
 Windows Vista Home Premium SP1

Sensors Data:

*Note: This values were obteined while running an Arcade (MAME) emulator for an hour with no over-clocking for both the CPU and the GPU. Also note that the Bios was patched from A09 to A12 before the second screenshot. Also No paste was aplied to the ACPI but it was afected by the new temps of it´s neighbors.

Before:


After:


Graphs:

(Minimun Temps)
This values of Temperature
were taken after resuming
the system from hibernation



(Maximun Temps)


T This temps were taken after
playing for half an hour
The game:
Never Winter Nights 2
average-high settings




Notes:

 The laptop is cooler in the exterior
 It cools faster in a 30 % - 40 % less time


Ok he also asked me to thank all the users of this thread he is really happy with these results, since his m1530 is really hot :S.

 

It seems like the drop in temp due to IC7 also lowered the temp of the hard drive! Unless that was just from the new bios haha.

 

that last test set looks odd.

better tim usually can leave idle tems near the same as they used to be, as less thermal disipation is needed, but load temps are where you will see the large difference.

but that set had huge idle changes and almost nothing for load, and that just does not make sense.

i guess it has something to do with fan control, the fans must not kick on/off until a high temp range.

 

I also guessed that, I just posted what my friend sent me, perhaps those are the problems from the xps m1530 you know... bad cooling systems, bad heatsinks, loose contact to the heatsink, bad drivers, faulty nvidia core...

 

So is this stuff better then ac5 for replacing thermal pads?

 

HELL YEAH, take a deeper look at this thread.