Kitguru just tested seven thermal pastes; Arctic Silver 5 and Céramique 2, Cooler Master's MasterGel Pro, Cryorig CP15, EKWB Ectotherm and Thermal Grizzly Kryonaut. They wanted to see what performance difference, if any, existed between them for no matter how effective your cooler is, it can't dissipate heat that is not transferred to it from your CPU. Their test was conducted with a i7-4790K CPU and Cooler Master Hyper 212 LED CPU Cooler and the results show that the incumbent is not necessarily your best choice.
"Following on from our previous articles about fan configuration and static pressure vs airflow fans, today we are looking at thermal paste. Specifically, we are hoping to find out whether or not choosing different types of thermal paste actually makes any difference. To do this, we test 7 products from 6 companies to see how much difference thermal paste really makes."
Here are some more Cases & Cooling reviews from around the web:
- Alphacool Eisbaer 360 Liquid CPU Cooling System Review @ NikKTech
- be quiet! Pure Base 600 Chassis @ Kitguru
- Jonsbo UMX4 @ techPowerUp
- Phanteks Enthoo Luxe Tempered Glass Edition Review @ NikKTech
Can confirm. I was using
Can confirm. I was using Artic Silver 5 all these years, but I didn’t like the way it spreaded. So I got kryonaut and replaced my GTX 970 paste with it.
My temps went from throttling under load to no throttling with a slight overclock.
On my CPU temps fell 7c.
Based from comment by
Based from comment by der8auer, an Overclocker, apparently Thermal Grizzly’s Kryonaut is the common enthusiast TIM used by extreme overclockers.
For best usage to reach OC records, he suggest evenly spreading the paste across with the stick this tube comes with. Also, for optimum performance, he suggest to change it once a year.
However, if you prefer the bead method and changing the paste every 5 years, he says your temps won’t be as good, but won’t be bad either (~5C variant).
Best cooling I ever got was
Best cooling I ever got was back in the day (90’s) using Keen’s mustard for 14 days, using everclear 95% grain alcohol to clean then re-applying, wait 14 days then repeat, All to push a 66mhz prossessor up to 71mhz
THOSE DAYS FUCKING SUCKED AND I NEVER WANT TO GO BACK, FOR FUCKS SAKES, NEVER AGAIN!!!!!!!!!!!!
10/10 Would snort mustard
10/10 Would snort mustard powder again.
No Shin Etsu? They make the
No Shin Etsu? They make the best. Of course its not consumer garbage so they wouldnt test it.
The 25% of people who keep the default for Intel coolers and other well made coolers are smart. It usually lasts forever if its the pad.
Thermal pastes are mostly snake oil and bullshit. Which is why so many people DOWNGRADE PCs when they buy aftermarket junk.
They dont realize that thermal cycling for a few months destroys their consumer junk while the stock cooler and pad would be stable and reliable for YEARS.
Would you mind explaining any
Would you mind explaining any of the points you made?
You made quite a few claims without actually proving any of it.
Not being sarcastic, genuinely I’d like to know.
Sure. Thermal cycling is just
Sure. Thermal cycling is just hot/cold cycles. Most thermal pastes dont handle that well and eventually they lose their thermal conductivity from expansion and contraction or liquification and resolidification.
The mostly pre applied pads made by big chemical companies that consumer sites have never heard of are much better formulated for longevity, arent susceptible to thermal cycling and NEVER need reapplication.
I know that Intels Delta air coolers, Asetek(who makes many rebranders coolers like Corsair, Cryorig etc.) actually use these types of pads.
It infuriates me to read articles about how people wipe that GOOD stuff off, only to apply some overpriced shit that gives them a 2 degree lower temperature immediately, but very soon their temperatures rise higher than they’d ever get with the stock compound.
Those huge chemical companies make stuff for servers that have to last for years without reapplication. They dont use that consumer junk on supercomputers for a reason: its junk.
A lot of the aftermarket stuff can also chemically react with the IHS or heatsink/coldplate if its not designed for the specific metal its being used on too. That causes it to break down, as well as the thermal cycling, and need reapplication. Knowing what metal is being used beforehand is essential to a long lasting thermal interface.
Bottom line: whats on a well made(Delta, Asetek) cooler is probably infinitely better than some shit in a flashy tube.
Did you do an actual test?
Did you do an actual test?
Not personally, but i do read
Not personally, but i do read plenty of them. Why?
Because you make intriguing
Because you make intriguing points that defy the convention many people and maybe others would like to see the same tests, results, and evaluations you apparently have been privy to. So, please, if you have more specific info to share, by all means do.
Edit: “…many people
Edit: “…many people *have…”
Its not that im privy to it,
Its not that im privy to it, nor does it really defy convention amongst people who work with computers. It defies marketing nonsense directed at consumer grade electronics users.
The world of servers and supercomputers is vastly different from “gaming” or “enthusiast” hardware. The stuff i mentioned is basic knowledge and common practice.
Maybe you need to read up on
Maybe you need to read up on material science then as thermal fatigue only happens in materials that are either structurally weak or can’t flow fast enough to compensate for the stress induce via thermal cycling, the hints in the name, thermal “paste”
Maybe you need to because
Maybe you need to because everthing you just said there is irrelevant when the issue is that the cheap consumer market TIMs dont stay the same consistency that they are when theyre in the tube or freshly applied.
So basically you can pretend you know shit about materials science by googling and misuing some terms but you dont know how it works.
Seriously? You’re citing a 10
Seriously? You’re citing a 10 year old study of electronics.
Things have moved on a little in a decade you know and the study you linked to is utterly irrelevant as it’s a stress test designed specifically to test for TIM degradation over time so bad and good TIMs can be identified.
That doesn’t mean most thermal pastes don’t handle that well and eventually they lose their thermal conductivity, it means some do and some don’t.
First off, “seriously” this
First off, “seriously” this is a comments section, not a technical journal. Its not really a citation as much as the first article that was handy.
There are plenty of actual research papers written on the subject if you know how to use google. Maybe read them too?
Heres a clue: search for thermal interface pumpout.
The basic physics of the phenomenon hasnt changed and never will.
Also, you do realize that plenty of consumer tech gets WORSE over time because its all marketing bullshit and your dollar has been destroyed by cheap Chinese imports.
Take monitors and TVs. The worst thing to happen to them happened about 10 years ago with the advent of the W-LED backlight.
Consumer level monitors have stagnated and in many cases gotten significantly worse in the last decade in terms of backlight spectral power distribution, and therefore color reproduction. Meanwhile professional ones have steadily gotten better.
You used to be able to buy an RGB LED backlit TV or monitor for a little more than a WCG-CCFL. Now you cant buy either and most consumer grade monitors are absolute shit for color 10 years later.
So if you think some “enthusiast” marketibg based consumer junk is better than well researched TIM, use it and enjoy reapplying it every year or whatever.
Yea and an article that
Yea and an article that doesn’t even support the original hypothesis, it’s an article about a testing methodology not an article on how thermal grease degrades over time or how pump-out effects real word usage.
Hears a clue, the internet isn’t full of people complaining how their CPU/GPU temperatures are increasing over time and you know why, because the degradations so small you’d be lucky to see a single ° change in a decade.
It’s not that thermal grease doesn’t degrade over time it’s that it degrades so slowly that it’s all but meaningless.
If you bothered to google or
If you bothered to google or read anything youd realize that pump out and drying can occur pretty quickly. Depends on the formula of the TIM in question.
Then instead of saying “Most
Then instead of saying “Most thermal pastes dont handle that well and eventually they lose their thermal conductivity” you should have said “depending on the formula thermal pastes can eventually lose some thermal conductivity”
But then again i suspect that wouldn’t have gotten the reaction you were looking for as people would have just laughed at you for stating the beelin obvious.
I should have said “most of
I should have said “most of the shitty consumer snake oil ones”.
Why do I ask whether you’ve
Why do I ask whether you’ve done testing? You seriously can’t think of a reason why?
Without proper evidence all I’m reading are unsubstantiated claims.
What the fuck difference does
What the fuck difference does it make if i personally test it or read someone elses research?
Do you PERSONALLY test every part of every product you use in your daily life? Thats a good one!
Did you just go off on me
Did you just go off on me over thermal paste? Not sure whether to laugh or to cry. I haven’t met anyone as childish as you in recent memory.
lf> your research
"The strain of anti-intellectualism has been a constant thread winding its way through our political and cultural life, nurtured by the false notion that democracy means that 'my ignorance is just as good as your knowledge."
― Isaac Asimov
Oh yeah youre one of those
Oh yeah youre one of those people who doest know what Rayleigh scattering is but if someone says “the sky is blue”. Your reply is “citation needed!” because you get your information from bullshit sources.
Try reading like i said.
Uh actually, you don’t know
Uh actually, you don’t know anything about me. Also, you’re acting like I read your first reply and replied and this is a follow up. You posted twice, I’m not sure what the hell you’re expecting me to read. Why don’t you re-read the entire conversation from the beginning and see why your followup comment makes no sense.
You act as if the sky can look blue is as obvious as having so many thermal paste be junk which is not the case. You are equivocating.
What makes you think I get information from bullshit sources? Like, where did THAT come from? Are you trying to convince people or are you here to troll? I’m asking a question I know the answer to. Whatever credibility you acted like you deserved you’ve now lost.
one of the best products
one of the best products ever: