Best long lasting Thermal compound?

So I have tested so many thermal compounds lately it's basicially insane, I am trying to find a thermal compound that not only performs well (a lot do extremely well) but also lasts. I experience on mmy CPU that the thermal paste tends to pump out after a while. I tried the following thermal compound only to have reduced performance after a couple of months:
- GC-Gelid
- Kryonaut
- Coolermaster HTK 3 (its a cheap one)
- Arctic MX 2 & 4
- IC Diamond
- Noctua NT-H1

I ordered coolermaster Mastergel, which will hopefully get here by friday and see how that goes.

I did a 10minutes aida64 test and the temperature is basicially abysmal given my notebooks cooling solution, this is the Prolimatech Pk 3 after ~1.5months (temps shouldn't be this high):


I hope the mastergel will do a better job. So far only conductonaut/LMU was able to stay at the same performance for months without degrading.

I'll keep you updated.

 

Conductonaut is pretty stable but it might be worth adding some washers if you notice it pumping or aerating too much.

 

Thermal Grizzly Conductonaut(LM) on 6820hk(after about 1 month) in ms-1785 with fans speed on auto:

 

Wow, that's silent mode indeed, a little hot for a GT 72. What temps you getting on agressive fans?
also how much watts does your CPU need when running aida64?

 

Here you go, fans with full speed:

 

Arent you able to see the wattage of your CPU in HWinfo?

But yeah, thats expected of an MSI notebook with liquid metal and full fan
Whats the highest clock you could manage on the notebook?

 

Then what are you doing wrong with ICD7? I repasted a G73JH, and I got the same level of cooling after 5 years of usage off the same paste job, never redone. My 1710 was repasted a year ago, and it still performs as well.

If your ICD7 job is degrading, then something is wrong.

 

It is just pumping out somehow.

 

When I apply it, I just dab it in the middle, about the size of a large pea, and I don't spread it. With ICD7, it's ok to put too much because it won't lose thermal capacity, and it's non-conductive. Just plop in the middle and attach heatsink, and let the thermal heat spread the TIM.

IIRC, you like to spread your TIMs. Big no-no for ICD7.

For rectangular chips, a line lengthwise 3/4 the length of the chip works well. No spreading.

 

Not quite, I never spread with ICD because i can't, it's way to thick. Also I do just put a line in the middle through the GPU die.

 

http://forum.notebookreview.com/threads/thermal-compound-test.814741/page-2#post-10698435


"Never tried to be honest, I just like to spread it since it makes the least amount of mess. But yeah for all people care you can just use pea method and be over with it. Performance is the same either way, you'll have to do way more cleaning when repasting tho."

You mean you have changed how to apply thermal paste?

 

@Danishblunt

I use the X Cross method with ICD as the pea method doesn't spread well enough due to how thick it is and spreading it will create a mess when you apply the heatsink on as it will splurge out.

The X Cross method is the best in my experience as once the heatsink is fitted, it provides the maximum coverage with the least air bubbles

 

hahaha harrrrsh! the internet / papusan never forget!

 

The spreading isn't that much of an issue, it's just pumping that it happening.

 

I'm a bit confused. Are you spreading TIM (from one post) or using the line method (or in another)? Or are you saying you use different methods depending on the TIM?

In general, with standard TIM (i.e. non-gallium based), a pea, cross, or line, one would probably have more z-axis (height) that may accommodate a poorly fitted heat sink better than if it is spread and there is not ample height to bridge the gap between CPU's IHS and the heat sink.

 

Whern it's not IDCDiamond I spread it, when it's ICDIamond i just "draw a line"

 

@Papusan boy, are you harsh

If the temps are good, it's all good. I don't mind a little run off, since I won't be seeing it.

 

If the heatsink allows it (think EM/SM series heatsink or vapor chamber) I always press down on the heatsink as I screw it in to spread paste and ensure the pads make contact.

 

Autofan with Mastergel:


So far rather imperssed

 

Autofan?
We are not able to see the speed of the fans during the test.
Through the software from the fan speed control you can set the maximum speed for low temperatures and call it "autofan".

Another thing:



"another one":

According to information from hwbot.org on your profile(https://valid.x86.fr/fbp8bl), you are using
Clevo P370SM-A. Following the information on this page, you used 4 drives on your laptop at the same time.

Samsung SSD 850 EVO 500GB
Crucial_CT240M500SSD1
Samsung SSD 840 EVO 500GB mSATA(1)
Samsung SSD 840 EVO 250GB mSATA(2)

This information keeps us convinced that the disk Crucial_CT240M500SSD1 is Sata and NOT mSata drive like for example Crucial CT240M500SS D3, which is very important.
Sata disk location on your laptop:



My MS-1785(msi gt72vr barebone)and my 970evo location:


My 970evo with heatsink having its own fan that allows low temperatures(like 28°) to be reached when it is turned on:




How did you achieve a decrease of 8 degrees Celsius on the SATA disc hidden behind the lid without using external cooling and with cpu/gpu fans (set to auto) that do not have any contact with the sata bay so they can not affect its temperature???

Of course, I also have in my mind the fan mod which has nothing to do with SSD drive temperatures anyway. And the best one, my favorite thread: P375SM-A Vortex 1070 and the "mystical modification of the heatsink"...
which has been done using technologies unknown to us and extraterrestrial knowledge that you do not inform us about because you must:

Just take a few photos of the cooling system and write a word about modifications, it takes less than 10 minutes. Unless you have something to hide and you are afraid of checking the correctness of the results you present, by members with the same or similar computer equipment.
Eleven days ago i asked you for answers, you said you are to busy. And now i see you are active in "hundreds" threads. So now you have a time then show to us the heatsink and make a description of the modification.

Show some respect to others by providing specific answers to the question and not avoiding them by finding excuses.
If I'm wrong, I'll apologize to you, but for now, in my eyes, you're not a trustworthy man.
Sincerely

 

Note that 2.5" sata drives have their case as a heatsink.

 

I'm not saying that they dont. I wanted to prove that external cooling was used for the test with Mastergel thermal compound, by comparing the temperature of the SSDs. My samsung 970EVO starts and stay @ 37° same like his CT240M500SSD1 in first tests, and suddenly in the second test the temperature of his SSD is a few degrees lower on "autofan"... when my 970evo required using external heatsink with built-in fan to get similar results.
So how could internal fans responsible for the processor and graphics card reduce the temperature of his SSD by 8° having in mind the location of this SSD in his laptop?

 

The 2.5 inch drives do have vents below, the fan will cause some incidental airflow, I dont know the exact amount of incidental airflow caused however.

 

Let me point out why you reply is nothing but spam at best.

By autofan I mean EC controlled Fan from the Clevo P375SM-A. You'd have to be awfully stupid to not understand this, since Clevo has no fan control to speak of, unless you purchase one from Obsidian, for testing purpose I shutdown the fan control in both tests to have a fair representation of the temperatures.

Yes the SSD is quite a distance away from everything. A SSD is getting warmer the more it's being used. *hint* *hint*

Now if you had asked me what i did before benching, I would have said:
-> First Becnhmark with older paste that started to pump out after a while. I started to notice higher temps in overwatch and witcher 3 (both games on my crucial SSD), so right after playing overwatch I started to benchmark.
-> Second Becnhmark I repasted my notebook and started the benchhmark right after i started my notebook.

My OS is installed on my 250GB Msata SSD, so in that benchmark the Crucial did nothing but idle. While your SSD is probably the one you use for OS. So obviously yours should be way hotter than mine.

Also comparing your M.2 SSd to my crappy crucial SATA SSD in terms of temps just shows you have no clue what you're talking about:


Even M2 drives differ from temperature depending on how it's built.

 

A new day, new hopes, new opportunities for me to learn something new,
let's get started!

Step 1.

The first test, your CRUCIAL SSD reaches 37 degrees.
The second test, your CRUCIAL SSD reaches 28 degrees.
Idle with 9° differential Hmmm?

AIDA64 program with standard settings shows only the "system drive" temperatures, not additional! I just checked it on win7 and win10.
In order for the program to show only the temperature of the additional drive, you need to deselect the system drive in the advanced options and select the additional drive. So if i can be honest, it looks like a total lie to me, but I'm happy to find out what the truth is, which you will probably present to us.


Step 2.

Very "constant" disk temperature result. I also started the test immediately after turning on the notebook (which was not used for several hours).
The temperature of both drives showed increasing tendencies(OUH! and btw this time I used two drives to check something I will write about in step 4).
Both drives show increasing tendencies if we talk about temperature.
Crucial(system) start at 25° get to 33°,
Samsung start at 25° get to 34° during this 10 min test.




Step 3.

A very uninteresting movie. We compare the temperature of the drives with the CPU load, not the temperature when copying files that are over 100GB.

Step 4.

It is a very interesting and motivating opinion, but let's check what the result came out in the tests.
I was inspired by the fact that a non-system(additional) drive can reach such low temperatures when it is not used, so this time two units were used for tests.
Crucial_CT256M550SSD1 Sata
Samsung SSD 970 EVO 500GB M2
To avoid unnecessary searching for a hole in the whole (as we like to say), the disks were used alternately and with different systems.

Test 1.
Win7
Crucial_CT256M550SSD1 - primary(OS drive)
Samsung SSD 970 EVO 500GB - second(idle one)


Crucial 36°-36°
Samsung 37°-37°
Hmmm... ok let's see another one!

Test 2.
Win10
Crucial_CT256M550SSD1 - second(idle one)
Samsung SSD 970 EVO 500GB - primary(OS drive)


Crucial 35°-36°
Samsung 36°-37°
Hmm... but where is my 28° on IDLE-ing second drive? what did I do wrong?

Is Sata driver as second drive getting lower temperatures?
the answer is - f***ing definitely not!
Is M2 driver as second drive getting lower temperatures?
the answer is - f***ing definitely not!


Summing up my entire post, dear colleagues,
please pay attention again to the temperature of Danishblunt crucial drive from the first post of this thread, and compare it to the temperature which I presented to you in test 1 and test 2.
We both used crucial drives, and please do not fantasize about which controller heats up more.

Cheers.

 

@ kothletino
Sorry can't be bothered to read more than step 1 because it's already full of garbage.



There is your proof that OS runs on my Msata drive.

 

Dear Danishblunt,
I installed AIDA64 program on two systems, window7 and windows10. In both cases the only disk whose temperature was shown(by default) was the system disk. And in your case it showed a completely different disc, why? You have change default settings because of what reason? And why temperature of only 1 of 4 drives?

At the very beginning I thought: "maybe I'm wrong and this guy lives on the pole where it's cold and maybe that's why these temperatures are so low". But the system language in the screen that you showed says: Denmark, possibly North Germany, so it isn't north/south pole.
In my previous post you got clear proofs of the temperature of both system and additional disks in IDLE during the test (Sata and m2, and your theory about the second drive in IDLE does not necessarily work).
I have performed a series of tests after which I am able to present hard evidence.
Maybe now you try to prove to me that you do not use external cooling or fans@max on your laptop while testing.
The temperature results of your tests are unrealistic with factory cooling at auto. At the moment I do not have any hard evidence from you that it is different and that means that your results are not quite true.

Because I see your activity again, I remind you about the cooling modification for 1070 in the thread P375SM-A Vortex 1070. Remove the laptop cover, take a picture of heatsink, write 3 sentences about what you have modified, that 120w or 150w gtx1070 work with high 60°s to low 70°s temperature. My 80w gtx 1060vortex during gaming stay at 75°(fans with factory settings) I am just jealous and I would like to have such low temperatures as well. It will take 3 min, Please.
Or that test was also done with autofan ouhhhh i mean FN+1(9)= max fan speed? If so then do not bother yourself, I know enough.

In the next tests, just mention the rpm of the fans, which will allow others to better understand the results and avoid unnecessary work.

 

1.) How would I know, I didn't write the program. All I can tell you is, as you have seen on my picture, my OS is indeed on my Msata drive.
2.) You temps are irrelevant, you should know that. Every drive has different temps which you could see on that youtube video as well. Maeker provided an explanation as to why my temps are better.
3.) You see, the more self entitled people are the less I care.

 

Maybe in diffrent way.
mastergel has reached temperatures of 55-58° in cpu stress test.
In that case, how the temperature looks with a Thermal Grizzly Conductonaut in the same cpu stress test?

 

When freshly applied the same. If you read the thread it's not about the thermal pastes performance when applied, but how long it lasts.
On fresh repastes:
- LMU
- Conductonaut
- GC Gelid
- Kryonaut
- TK3
- ICD
- CM Mastergel

All perform about the same. There are some degrees difference in some applications, but the basic performance is around the same. So far Mastergel does spread out very differently than the other pastes. I don't know how to describe it, it looks like the paste is trying to bond the heatsink with the CPU instead of it being mashed together.

 

Sorry for offtopic, but good mod is to throw away master GPU's stock fan and replace it or mod existing one with P870DM fan.
http://forum.notebookreview.com/threads/better-gpu-fan.824813/

 

Still w8ing for it.

 

What was the total price with delivery fees?

 

I ordered 2 items, the total price was almost 26€.

 

I'm really going crazy, what was this thread aiming? SSD Temps or Thermal paste performance after a very long time?
DanishBlunt You did mention about thermal performance after using CM Nano in Best Thermal paste thread as well? Do you use same Clevo with Stock cooling or custom cooling?

 

Would these fans be compatible with the P170SM ?

 

Some modding is needed but you can make it work. We use basicly same hwatsinks.
You either replace the motor and fan so you can use old fan housing and same mounts.
Or you grind new fan shorter about 10mm and use foil tape to secure it to the heatsink.

 

Excellent! Thank you!

 

idk what the guy was thinking either with my SSD temps.

my cooling is modded, tho atm in a "half finished" state. Also no, I mentioned in best paste "CM HTK", nano is an actual first timer for me.

 

Half finished? Do you plan on water cooling mod?
I thought the temps were after using CM Nano gel.

 

The temps are after using CM nano gel.

No, vapor chamber. Needed some tools to finish it, hopefully soon coming from china.

 

Hmmm.

 

This guy proved that you used external cooling in CPU temperature tests by comparing SSD drive temperatures.
I asked the owner of clevo p375 how the temperature of the additional disk on which the system is not installed and which is not in use are, his answer is:

"The only situation when the SSD in IDLE reaches 28°C is one where the drive sits on the fan which constantly blows him with air."

Are we clear or you still don't understand what i mean?

 

Or something else. It's called ambient temperature.

 

What's your ambient temps at the time of test?

 

Being in a colder environment or using some chilling is an external factor, you can't just ignore it when making comparisons.

 

So even a climate change is taken as external factor?

 

Lol in the sense it impacts ambient then yes.

 

Where have you been buying the pastes from? aliexpress or other chinese sites?

Even though they claim to be genuine dealers, almost everything is fake or refurbished.

i am a long time user of MX4. Use it in my desktop laptop playstation xbox, and havent had that much of a problem. i buy it from arctic cooling website.

 

usually danish or german sites, so extremely unlikely fake or refurb.