[Boards: 3 / a / aco / adv / an / asp / b / biz / c / cgl / ck / cm / co / d / diy / e / fa / fit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mu / n / news / o / out / p / po / pol / qa / r / r9k / s / s4s / sci / soc / sp / t / tg / toy / trash / trv / tv / u / v / vg / vp / vr / w / wg / wsg / wsr / x / y ] [Home]
4chanarchives logo
How does /g/ apply thermal paste to the CPU?
Images are sometimes not shown due to bandwidth/network limitations. Refreshing the page usually helps.

You are currently reading a thread in /g/ - Technology

Thread replies: 200
Thread images: 38
File: fO-tGPxg9EIu.878x0.Z-Z96KYq[1].jpg (40 KB, 405x323) Image search: [Google]
fO-tGPxg9EIu.878x0.Z-Z96KYq[1].jpg
40 KB, 405x323
How does /g/ apply thermal paste to the CPU?
>>
>>51407478
Hot
>>
File: sSnyqcL.jpg (883 KB, 2736x2052) Image search: [Google]
sSnyqcL.jpg
883 KB, 2736x2052
>>51407478
Like This.
>>
File: 1444393627918.png (834 KB, 789x859) Image search: [Google]
1444393627918.png
834 KB, 789x859
>>51407478
put that shit all over my dick and mushroom stamp that CPU right in the FUCKING SHIELD BB
>>
Carefully.
>>
File: image.jpg (2 MB, 3264x2448) Image search: [Google]
image.jpg
2 MB, 3264x2448
Check em
>>
Pea in the middle
>>
File: 1441014888470.jpg (312 KB, 1042x860) Image search: [Google]
1441014888470.jpg
312 KB, 1042x860
>>
I usually apply it with a butter knife.
>>
OP, it doesn't really matter.

I just do a little glob about the shape/size of an uncooked grain of rice. As long as you don't put way too much on, you will be fine. Which technique you use results in very minimal differences.
>>
File: MSI.webm (3 MB, 1280x720) Image search: [Google]
MSI.webm
3 MB, 1280x720
like this
>>
File: excellent.gif (32 KB, 650x523) Image search: [Google]
excellent.gif
32 KB, 650x523
>>51407700
double dubs confirm
this is the one true way
>>
File: 1435851845703.gif (4 MB, 298x224) Image search: [Google]
1435851845703.gif
4 MB, 298x224
>>51407700
I thought this was a legit MSI Webm at first.

Almost fucking had me.
>>
>>51407700
The best part is that you can tell by the way the person is scraping it out that he isn't sure what he is doing.
>>
>>51407700
oh god im cringing so hard right now
>>
>>51407774
Sorry to tell you this...
It is.
>>
Some pasta

>got a 10 year old PC at a flea market for $5
>it works
>decide to do all kinds of cruel experiments on it because I can't think of anything else to use it for
>remove CPU cooler and see how long it will run before frying
>it reaches 112 degrees celsius but doesn't fry
>as a reward I test the rumor that semen is just as good as thermal paste when putting cooler back
>but instead of cumming in a cup and applying the semen with a brush as recommended I just take CPU out and cum directly on it
>it works, temps only go up to 80 degrees during next test
>wipe the HDD, install xubuntu, and backup 70GB of hentai on it
>keep it in a closet for a couple years
>no longer need it as backup because I got two dozen 256GB microSD cards for free thanks to /tech/
>sell it on ebay
>someone bought it for $8
>some faggot unknowingly bought a PC with a HDD stuffed to capacity with futa doujins and a hot load of my semen stuck between the CPU and heatsink
>>
>>51407478
We use sperm. Cum of black men work better
>>
A pea sized dot then squish it down with the heatsink.
>>
>>51407478
X shape. It works better.
>>
>>51407700
THAT BLOB ON THE EDGE
>>
Wearing a bikini
>>
>>51407869
>>got a 10 year old PC at a flea market for $5
>>it works
>>decide to do all kinds of cruel experiments on it because I can't think of anything else to use it for
>>remove CPU cooler and see how long it will run before frying
>>it reaches 112 degrees celsius but doesn't fry
>>as a reward I test the rumor that semen is just as good as thermal paste when putting cooler back
>>but instead of cumming in a cup and applying the semen with a brush as recommended I just take CPU out and cum directly on it
>>it works, temps only go up to 80 degrees during next test
>>wipe the HDD, install xubuntu, and backup 70GB of hentai on it
>>keep it in a closet for a couple years
>>no longer need it as backup because I got two dozen 256GB microSD cards for free thanks to /tech/
>>sell it on ebay
>>someone bought it for $8
>>some faggot unknowingly bought a PC with a HDD stuffed to capacity with futa doujins and a hot load of my semen stuck between the CPU and heatsink
If this is real,then good god I laughed like a hyena
>>
LOL! Truly a great meme!
>>
File: ohshit.jpg (79 KB, 655x401) Image search: [Google]
ohshit.jpg
79 KB, 655x401
>>51407478
Very frantically.
>>
>>51407520
I don't feel good
>>
https://www.youtube.com/watch?v=TnyoJtv9Cx0
>>
File: 1445264698444.jpg (118 KB, 497x640) Image search: [Google]
1445264698444.jpg
118 KB, 497x640
>>51408124
>some faggot unknowingly bought a PC with a HDD stuffed to capacity with futa doujins and a hot load of my semen stuck between the CPU and heatsink

top fucking kek
>>
File: 1389096193693.gif (3 MB, 235x320) Image search: [Google]
1389096193693.gif
3 MB, 235x320
>>51407869
holy shit
>>
>>51407869
Best reply ever.
>>
>>51407869
fucking kek
>>
wait you're supposed to apply thermal paste?
I thought it was a /g/ meme
>>
>>51407478

like im knocking one out
i just let it spurt everywhere

im getting a semi just talking about it now
>>
Pea/Dot method.

I tried spreading it manually once but the fucking paste dried up as soon as it touched the IHS so I couldn't fucking spread it.
>>
File: 1443534533436.jpg (439 KB, 2048x1536) Image search: [Google]
1443534533436.jpg
439 KB, 2048x1536
>>51407478
>>
>>51409529
nice and cool
>>
I take the computer to a professional store where a qualified technician will do it for me.
>>
>>51407869

I'm never buying anything on ebay now
>>
File: 1444032080004.gif (2 MB, 236x224) Image search: [Google]
1444032080004.gif
2 MB, 236x224
>>51409627
>paying someone to do something any idiot can do.
>>
>>51407573
>>51409627
This is what you get
>>
>>51409627
I bought a heatsink from a computer store, and they said they could install it for like $5 extra. Sure, why not.

They took an hour to install it, and while I browsed the shop I would occasionally look at the "technicians" working on my computer. It looked like they were staring at some kind of alien machine, they had no idea of what the fuck to do.
>>
File: 1435192151500.jpg (35 KB, 245x274) Image search: [Google]
1435192151500.jpg
35 KB, 245x274
I don't know. I'm about to install my first CPU and I'm scared I'll fuck it up.
>>
spread method, fuck everyone
>>
>>51410022
me too, anon, me too
>>
>>51410022

I'm going to get my first actual PC soon and I'm terrified of putting it together myself cause I'm sure I'll break something

The online store I'm picking the parts from can build the thing for an extra 45 bucks and I'm actually considering checking that option even if it makes me a total pleb
>>
>>51410022
>>51410064
>>51410066
Get an old PC, smash it to smithereens and then build it back again. You won't lose money or burn a brand-new 19 thousand dollars CPU in the process, and you'll learn basically the same.
>>
>>51410066
I've heard it's like playing adult Legos. I can follow instructions and connect cables and sockets, so I'm not too nervous. But I guess I'll find out.

The only things I'm unsure of are applying the thermal paste and installing Windows from the BIOS.
>>
>>51407643
wow a piece of plastic

would have loved to be in the marketing meeting for this one
>>
>>51410124
installing windows isn't hard. the thermal is the difficult part...
>>
>>51407774
this is a real video anon
>>
>>51410099
>>51410140
Is this man's method sound?

https://www.youtube.com/watch?v=rjF5jabXRCY
>>
>>51410210
My english sucks... BUT THIS GUY'S, HOLY FUCK.
>>
>>51410210
i'm not an expert, but this guy doesn't know what he is doing also
>cooking on pc
>>
>tiny drop of thermal paste the size of a pea in the middle of the cpu
>slowly but firmly press the cooler on top of it, straight down without twisting
>done.jpg

seriously, how fucking hard is it to apply a layer of paste? and who the fuck spreads out their termal paste?
>>
Don't spread it out, let the cooler spread it
>>
>>51410296
I spread it out. It guarantees coverage on all the surface.
>>
>>51407869
You, if this is real, I swear. I thought 8$ was a
Oh my god what the fuck dude are you fucking weird yeah you fucking are actually like what the hell my gentoo memestation my memestation has cum in it. I did format the drive though I'm not retarded
>>
it doesn't fucking matter which method you use, whether you spread it or not. just do the right amount for the respective method and you're good. your temps will be fine.

like seriously, i never had problems stemming from this part of putting together a pc. you gotta worry about faulty hardware more than the fucking thermal paste.
>>
>>51407478
>thermal paste
>not just cumming on your cpu to save shekels
>>
>>51410369
spreading is fine. it doesn't matter. spread method if done right gives excellent temps as well.

it. does. not. fucking. matter.
>>
>>51410210
This video shows how NOT to apply paste
>>
thermal paste is the biggest meme in the universe. all you do is put a little dab on and then put the heatsink on. that's it, you don't draw a shape, you don't use a card to spread it around. the act of putting on the heatsink spreads it to all of the area of contact
>>
File: 1383503379366.jpg (1 MB, 2048x1536) Image search: [Google]
1383503379366.jpg
1 MB, 2048x1536
>>
>>51410210
>proseccor
>>
File: IMG_20151018_102633.jpg (726 KB, 1532x1532) Image search: [Google]
IMG_20151018_102633.jpg
726 KB, 1532x1532
I have an AMD FX 8120 running at 5.3 gigahertz cooling this thing is hard to say the least 620 watts is crazy to cool I also have a 5820k @4.95. Applying thermal paste using different techniques definitely makes a difference it usually depends on the viscosity of paste you're using paired with a specific technique over the last 25 years I have seen it done a fuck load of different ways but the best way is to jest try different ways till you get the temperature or stability you want my 5820 K was not stable above 4.5Ghz using the center dot Method I tried several different ways and found using more thermal paste actually made it cooler and more stable MX3 FTW
>>
>>51410843
Here, you forgot these: .........,,,,,,
>>
>>51409627
>Friend's PC is breaking or something and somehow come to the decision that the heatsink needs reseating.
>His dad refuses to let me do it and instead gets it taken to local computer shop.
>The "tech" there wipes of the old gunk, applies new shitty silicon paste.
>Then proceeds to spread it with his bare fingers.
>I say mention that's not a good idea.
>He says it's fine and carries on.

You're probably more suited than most technicians.
>>
>>51410945
I can't be bothered with that shit I'm in my car driving and speech to text doesn't always do punctuation correctly also go back to reddit
>>
>>51410843
What exactly do you have that is pulling 620w?
Your GPU and CPU? Come on...
>>
>>51410022
as someone that built his first PC a year ago, there's nothing to fear, really. It's scary because it's an expensive thing, but just place it as the little arrow tells you and let it sit. When placing you don't have to put any pressure.

When you have to turn the little lever to properly seat it, remember that it's clamping that mofo down so that it never moves while it's working, so if you have to put some pressure, don't be scared.

The heatsink may be a bit of a pain in the ass if it's an xbox huge aftermarket one, but for the stock one it's also simple as hell. It probably comes with thermal paste pre applied so you don't even have to worry about that.
>>
The X marks the spot.
>>
I always use an x-pattern for AMD and 4+core i7 CPUs. Most of Intel's CPU only needs a thin rice rain-sized blob right in the middle running up-down. You need to get the paste right over where the actual chip and iGPU is.
>>
>>51407478
All my experience so far tells me that it really doesn't fucking matter. Use very little paste and don't spread, as those things won't make any difference at all. I even went without thermal paste once, just to see what happens, and very surprisingly, nothing significant happens. The CPU was literally the same temperature on idle. I guess if the surface of your cooler/CPU is very uneven, then it makes a significant difference, but otherwise it's not that important.
>>
File: 20151018_101123.jpg (3 MB, 5312x2988) Image search: [Google]
20151018_101123.jpg
3 MB, 5312x2988
>>51410978
The whole amd fx8120 pc with out a gpu with 6 140mm raid fans that draw 26 watts total at full load read on a kilowatt wall moniter gpu is powered by secondary PSU to get a more accurate reading
>>
>>51410976
>shitposting while driving

I pray to god you're the only one that dies when you wreck.
>>
File: 20150621_135350.jpg (1 MB, 3264x1836) Image search: [Google]
20150621_135350.jpg
1 MB, 3264x1836
>>51412085
Lapping does help displace thermal energy quite a bit
>>
>>51408769
But why would you reply to a reply that quoted the whole post? Why did you not reply to the original post?
>>
>>51412626
>not able to drive and talk at same time
>>
File: ctsnjIU(2).webm (1 MB, 720x404) Image search: [Google]
ctsnjIU(2).webm
1 MB, 720x404
>>51412626
Must be one of the plebs who think driving is hard
>>
>>51412629
You can also see where the Nvidia silkscreen that's printed on the die surface indented on the cooler
>>
>>51412085
Thermal temperature on idle won't change much, as heat transfer is less effective the lower the temperatures involved are.

In short - the greater temperature difference between two things, the more heat travels from hot to cold. There isn't enough of a thermal delta at idle for it to make a difference, but at load, the difference will be significant - especially with regards to certain cores being hotter than others.
>>
Which thermal paste should I use?

Are ear-sticks a good method to clean old thermal paste?
>>
>>51412802
Nope, can't confirm this. Went two weeks without paste (mainly because the tube I had dried up), and even on load, the temperature wasn't significantly higher than with paste. I didn't take enough data for proper statistics, but it really was pretty surprising to me.

By the way, in a more rigorous sense, the effects you describe do not exist. Idle temperature is pretty strictly proportional to load temperature. I leave the proof as a simple exercise.
>>
>>51413029
So, your one experience with CPU temperatures automatically invalidates the proof and statistics of hundreds of overclockers and PC enthusiasts over the past decade?
Men...
>>
File: images(4).jpg (7 KB, 304x166) Image search: [Google]
images(4).jpg
7 KB, 304x166
>>51410022
>>51410064
Corsair H100 comes pre-pasted
>>
>>51413726
does that paste expire/go bad?
i have sitting for two years, is it safe to use still?
>>
If the bottom of the cooler is perfectly flat put a small dot in the middle and press the cooler down on top of it.

If the bottom of the cooler is not flat (pic related), then spread it with your finger in a ziplock bag and/or plastic card
>>
>>51413737

If you never removed the block from the cpu it's not an issue, if you remove it you may as well refresh it. Clean it with alcohol and apply again.
>>
>>51413737
>>51413824

Forgot to specify, it does apply to any type of paste. H100 use Arctic 5.
>>
Exactly as pictured. One pea sized blob right in the center.
>>
>>51413786
If the bottom of the heatsink isn't flat, you apply the thermal paste in lines ON the cooler, and then apply the heatsink to the IHS.

Seriously, dude. What the fuck. How is this not obvious? If the goal is to use the heatsink to make thermal paste spread evenly, then you use the fuckin' heatsink.
>>
>>51413840
I thought Corsair used some weird Dupont paste
>>
>>51413726
so does stock intel
>>
>>51413737
Any thermal paste eventually goes bad because they dry out.

A notable exception are the liquid metal pastes, because those are completely different from average consumer pastes.
>>
File: IMG_0757.jpg (1 MB, 3264x2448) Image search: [Google]
IMG_0757.jpg
1 MB, 3264x2448
>>51413824
nah its still brand new, with the plastic cover over the motor/pump.

>>51413840
i have an H110

>>51413902
ffs
>>
File: IMG_0758.jpg (1 MB, 3264x2448) Image search: [Google]
IMG_0758.jpg
1 MB, 3264x2448
>>51413912
>>
>>51413912

Oh you mean you never used it and still have a sealed one ? I really don't know dude if tat thing dry out after some time. Better contact corsair!
>>
>>51413962
or i could go to micro center and buy some arctic paste and just replace it?

i honestly hate applying paste. wish someone would come up with a prepackaged strip that i can peel off and drop unto the cpu
>>
>>51414004
Or some kind of space age gel in a small tab shape, about the same size in every dimension as a CPU, maybe thinner, that's placed between the CPU and the heat sink, can be reused, makes the same thermal transfer as paste. That would be a godsend, I wouldn't loathe changing CPUs any more.
>>
>>51407520
/triggered
>>
>>51410066
>>51410064
>>51410022
First time is only scary because you make it out to be. Honestly, its like putting a puzzle piece into the right spot. Just be careful, hold your hand still, and lightly place it in where it says. My first time was real easy, got it in first try. I got tired of the stock cooler tho, bought a cm evo, and when i tried to pull off the old heatsink, the thermal paste solidified and i pulled the cpu right out of the socket. Didnt even undo the lever. None of the pins were broken, and that was a year and a half ago. Still running strong with my 8320 to this day. Just be clam and collective and youll be fine
>>
>>51407869
this was posted on reddit a couple days ago
>>
>>51407478
pretend the blue shit is paste.

this is my OCD way of doing it.
>>
>>51414049
yeeeeeeesssssss! im so tired of this 'use a pea size' shit, im spending $1000+ on equipment and you want me to guessimate how much thermal paste to use on what is the most important part of the build.
>>
>>51413029
The laws of thermodynamics don't exist?

In your anecdotal experience maybe not, but the effects I described are a fundamental property of physics. Heat transfer is proportional to the disparity in heat between two things - and entropy will always increase.

2nd law.
>>
>>51413029
However, you're not wrong in saying idle temperatures are proportional to load - this is correct. However, the effects of heat transfer are more pronounced with a higher temperature disparity (in this case, the very hot CPU, and the cooler heatsink. TIM = more thermal conductivity)
>>
>>51409902
>any idiot can apply thermal paste correctly.

Have you learned nothing from the horrifying pics in this thread?
>>
>>51414126
why do we need thermal paste
>>
File: amd-tim-6653410.jpg (310 KB, 1600x1200) Image search: [Google]
amd-tim-6653410.jpg
310 KB, 1600x1200
You don't need much. Remember it spreads.

IT pic related your ideal amount
>>
>>51414170
It evens out surface imperfections in the chip - increasing the uniformity of the transfer.

Lapping is a more precise form of this - it aims to create a perfectly smooth surface which will eliminate any "hot spots" that lack contact with the heatsink.
>>
>>51414194
so if i use >>51414174 amount and it doesnt reach near the edges does it matter?
>>
>>51414211
Basically, yes. Put a blob in the middle approximately the size of a pea, and the pressure from the heatsink will spread it when you place it on (the contact pressure is so high, it forces it over the surface of the entire chip - it's called capillary action)
>>
>>51414211
The CPU die is not the size of the entire thing. If you've ever seen a delidded CPU, that is the entire area that you need to cover with thermal paste; which is to say not even close to the whole thing.
>>
I use the X method

I've tried pea method and it was like 5c hotter
>>
>>51414280
Really? That's surprising. I'd be more inclined to believe it was more just bad luck or improper seating - the TIM's application is necessary, but the differing methods of doing it don't yield much difference.
>>
>>51414227
do you ever jiggle or twist the heatsink before clamping it in place?
>>
>>51414306
i tried 3 methods

line, pea and x

tried each of them 3 times

cooler was megahalems
>>
>>51414320
Nah - really no need. The compression will spread the fluid - a heatsink often has about 60lbs of pressure on it.
>>
>>51414323
Well, thanks for being empirical about it. I'll give the x method a try next time I seat my CPU.
>>
>>51407478
what happens if there is no themal paste or if there is only a little?

i've heard that the computer would attempt to boot for like 3-5 seconds then i would restart itself over and over again. Any truth to this?
>>
File: Untitled.png (7 KB, 437x364) Image search: [Google]
Untitled.png
7 KB, 437x364
>>51414350

red line is my X
yellow line is how people that does X and claims pea is better's X
>>
>>51414306
https://www.youtube.com/watch?v=EyXLu1Ms-q4
>>
File: Capture.jpg (74 KB, 930x654) Image search: [Google]
Capture.jpg
74 KB, 930x654
>>51414350

example of too long X
>>51414380
>>
>>51414366
see
>>51414227
without the paste there isnt a perfect contact, therefore more heat/uneven cooling = dead cpu
>>
>>51414380
Yeah, I imagined it was like this.

After all, the die in the centre is what needs to be covered.
>>
>>51407520

what the FUCKKKKK REEEEEEEEEEE
>>
File: Capture.jpg (79 KB, 914x662) Image search: [Google]
Capture.jpg
79 KB, 914x662
>>51414412
i think its because the pea method wont spread all over the heatspreader

basically the place that paste doesnt spread to is where the heatsink doesnt make contact

it's like having gigantic airbubble
>>
>>51407774
anon.. this is real and there is more than one
>>
>>51414459
Draining my loop tomorrow to move some stuff around so I'll give it a try.

Thanks anon
>>
>>51414401
wrong. a pea-sized glob of arctic silver is all that is required for optimal, even coverage. unless op is a faggot, in which case let him bake his cpu.
>>
>>51414484
imo
it needs to cover as much as possible

pea method only covers about 65% of the heatspreader

short x method can cover like 85%
>>
>>51407573
Can someone explain to a normie what is going in this photo.
>>
>>51414519
The same concept when you put 1,000 times too much jelly on a peanut butter jelly sandwich and bite into it.
>>
>>51414519
Need to spread it if paste directly on die.
>>
>>51414475
btw ive came to the conclusion that temp gets better the more coverage paste gets on the heatspreader

having airbubbles = less coverage
not spreading to the whole cpu and only a small circle in the center = less coverage
>>
>>51414400
shouldn't the heat help spreading it a bit more?
I agree that you shouln'd apply too little, though
>>
>>51414727
Depends on the paste on whether or not it needs to cure
>>
File: 5f5.jpg (36 KB, 638x477) Image search: [Google]
5f5.jpg
36 KB, 638x477
>>51407700
>>
>>51410350
see
>>51414785
>>
>>51414400
So, am I not supposed to use Arctic Silver?
>>
>>51414903
That was Arctic Silver 2

They have Arctic Silver 5 now which is better, but the best is Arctic MX-4 since it's carbon based and non conductive and doesn't need to be cured
>>
When should you reapply thermal paste?
>>
>>51415108
I like to do mine every 2 years or so

I recently changed out my paste to some good thermal compound after using the stock paste that came with my cooler and saw a 10-15C temp drop from the high 80s to 60-70C
>>
>>51414834
Bit the b8

This is a troll video from tek syndicate.
>>
>>51407478
Grain of rice was for flip chip processors. Pea sized is for heat spreader processors.
>>
>>51410969
It is fine. What's going to happen, the finger grease goblin is going to come and steal your mega hurts?
>>
>>51415108

When you remove the heatsink.
>>
>>51413912
>those monkey paws
>>
Done 4 different personal builds over 10 years and numerous shop repairs, dot/pea method has never let me down. Ever.
>>
>>51412715
This
>>
>>51416139
Really what I'd be worried about are two things:
1) That may cause air bubbles. I let the cooler spread it, from a dot in the center, to avoid that.

2) That shit's gross man, gets on f*cking everything. That means it got all over the outside even if he wiped up his hands before re-assembly / putting the case back together.
Eeeeewwww...

Also f*ck letting people touch your hardware.
>>
>>51407520
That's exactly how to treat an AMD cpu.
>>
>>51413902
I'd love to use some gallium as solid thermal paste. Unfortunately heatsink and heat spreader are usually made of aluminium which it dissolves. Still, funny thought.
>>
File: image_0.png (26 KB, 378x442) Image search: [Google]
image_0.png
26 KB, 378x442
>>51414081
>clam and collective
Clams confirmed for new world order
>>
>>51414100 what's gonna happening your pc fail safe triggers and shuts down. Truth is excessive amount of thermal paste will create a shittier thermal transfer rate.
>>
>>51414147
>>51414126
Load and idle temperatures are both solutions of the same poisson equation with different boundary conditions. If you do the math (literally), you'll see that load temperature is a linear function of the idle temperature, meaning if the idle temperatures in two configurations are the same, the load temperatures will be as well.
>>
>>51407478
>be me
>be 12ish years old
>building shit tier computer cuz older neighbor kid got me interested
>timeforthermalpaste.jpg
>ask how the fuck it works
>"you need to use sand paper on the cpu so the thermal paste can get as close to the cores as possible"
>"the thermal paste fills into parts that are scratched away"
>oh fuq didn't know that
>act like i know what i'm doing and sand the top of the cpu a bit
>tells me to use the credit card method for spreading
>pc just werks soon after getting the heatsink on
>mfw years later I realize he was trolling the fuck out of me
>mfw actually believing the credit card method isn't absolutely retarded
>>
Jesus christ and I thought on /g/ would be SOME clever techs.

You take a quarter of a pea sized spunk of thermal paste, put it on the heatsink (yes) and then you take a plastik bag, put it over your finger and rub the paste over the heatsink surface (that has contact with the cpu of course). Then you wipe it off with a clean cloth. You do this to have thermal paste in all the micro scratches and pitch of the metal piece to make it as even as possible to have as much contact as possible with the cpu.
After that you put the pea sized piece of thermal paste on the cpu and spread it even. I use the non-sharp side of a clean butter knife for this and it works just fine. The heatsink alone will *NOT* spread the thermal paste enough.
>>
>>51417689

That is the most autistic thing i have ever read. Biomass yourself IRL
>>
File: clock_boy.jpg (87 KB, 512x512) Image search: [Google]
clock_boy.jpg
87 KB, 512x512
>>51407478
>>
>>51417631
oh yeah, the credit card is the retarded part.
hahahahahaha omg
>>
The dot method sorta, but I stretch it out a little bit so it's somewhat a rectangle covering more of the die under the IHS. I found doing this helps balance the temperature of all cores. In my experience this was the difference between for example: 30c, 25c, 26c, 31c to 28c, 25c, 25c, 29c

I know this could just very well be margin of error though. But i seen a result, good enough for me.
>>
What is the go-to paste or grease to use now? I remember it being artic silver 5 but people say it's crap.
>>
>>51410022
>not having built your first pc with your dad when you were 4
>>
File: 1443964422175.png (1 MB, 1440x1014) Image search: [Google]
1443964422175.png
1 MB, 1440x1014
My homebrew method.
>>
File: thermalpaste.png (72 KB, 464x928) Image search: [Google]
thermalpaste.png
72 KB, 464x928
>>51418138
Cum.
>>
>>51418214

Test it again after 3 months
>>
>>51412715
driving's not hard.
If you're on an empty road with no unforeseen things happening, you don't really have to pay attention, yes.
BUT:
You don't know when the fuck in front of you is going to slam the brakes, or when a kid is going to jump out of nowhere, or when somebody is just going to pull out.
>>
>>51414459
I do what linus does

pea and a line
>>
>>51418138
arctic mx-4: good temps and price, non-conductive and lasts for years. suited for gpus and cpus (but not delidded ones).
>>
>>51418138
Artic MX-4 is meant to be pretty good but I haven't had much luck with it on laptops, at first its good and temps are low but it seems to go runny and pump out after 6 months making the temperatures high.

This has happened with several laptops which I ended up replacing with ceramique 2, 12 months on and everything is still good.
>>
>>51418214
apart from this thing being fake, how reliable is mayonaise actually?
>>
>>51418284

Even if it worked, it would smell horrible.
>>
>>51418284
It doesn't last long and oxidizes the CPU and heatsink http://www.hardwaresecrets.com/does-mayonnaise-last-as-a-thermal-compound/4/
>>
>>51418170
can someone post the full thread or the full pic of previous responses?
>>
>>51418275
>(but not delidded ones).
Would this explain my poor experience with MX-4 used in laptops?
>>
>>51418293
welp rip
>>
>>51418304
i believe so.
>>
>>51418320
Honestly I was really disappointed in MX-4 after being hyped up with good reviews, I even reapplied it a few times just incase. All heatsinks were cleaned with Arctic silver arcticlean and new lint free cloths.

Anyone else had any bad experiences with this MX-4 and laptops/delidded processors?
>>
>>51410350
>abit
This should be an improvement.
>>
>>51418345
just take a look at some oc community forums. they were 'forced' to delid many of the recent intel cpu models to get the results they desire since intel mostly stopped soldering the ihs. and they often don't like mx-4 for that task.
>>
>all cpus require the same method
>>
>>51412918
Clean it with a coffee filter.
>>
>>51418384
hmm I guess that confirms it then, MX-4 is piss poor for bare dies.
>>
>>51418250
Linus wants you to destroy your computer parts though
>>
>>51412918
Search for "Lint Free Chemical Resistant Cloth (28cm x 38cm) 20 Sheets" on ebay, I exclusively use them in my shop for repasting jobs.
>>
File: 1435936630956.jpg (8 KB, 250x232) Image search: [Google]
1435936630956.jpg
8 KB, 250x232
>>51407520
I did this with the mobo socket for intel cpus when I built my first PC, and then after realising my fuck up tried washing it out with water
>>
>>51407478
grain of rice
>>
File: 1440251017676.jpg (141 KB, 768x960) Image search: [Google]
1440251017676.jpg
141 KB, 768x960
>>51418694
Anon even children know you don't mix water and electronics.
>>
>>51412918
>ear-sticks
no no no no, there is cotton that will be left on the processor, use a coffeefilter or microfiber cloth
>>
>>51407869
>>no longer need it as backup because I got two dozen 256GB microSD cards for free thanks to /tech/
how?!
>>
A thin X with a dot in each sector.
>>
File: CPU_5.jpg (195 KB, 1000x884) Image search: [Google]
CPU_5.jpg
195 KB, 1000x884
>>51414211
yea, >>51414174
is pretty much perfect
and no, it doesn't need to reach the edges, the cpu itself doesn't cover the entire surface
>>
>>51410447
I saw the story that the other anon wrote...
But shit does this work
>>
>>51416273
>f*cking
>*
>********
what the fucking fuck is wrong with you, you fucking disgusting shit cunt
>>
>>51413900
>>51413726
and stock amd
they all do
they know the pictures in these threads and try to avoid fuckery
>hurr got an cpu an din du nuffin but is not werking
>amd/intel/corsair etc get their now broken products back and have to deal with a massive retard
>>
>>51418767
Thats actually a myth, just make sure you let it dry first.
it also works out better if you use distilled water but thats whatever really, Im sure normal water and then drying would be good enough.
iirc you can actually submerge a running motherboard in distilled water and it will be fine
>>
>>51422189
>you can actually submerge a running motherboard in distilled water and it will be fine
yea, for a short while

the problem isn't water itself, it's the things that dissolve in it that makes it conductive, plus it causes corrosion in some metals

submerging then drying most electronics in sufficiently pure water won't cause any damage, it's just not often that water-related cases involve both sufficiently pure water and/or sufficient drying out before the device is powered on
>>
>>51422303
makes sense that it would cause corosion, and yea thats why I said let it dry.. unreal how many people dunk their stuff and then try to "see if it works still" right away.
>>
>>51421708
>triggered
Thread replies: 200
Thread images: 38

banner
banner
[Boards: 3 / a / aco / adv / an / asp / b / biz / c / cgl / ck / cm / co / d / diy / e / fa / fit / g / gd / gif / h / hc / his / hm / hr / i / ic / int / jp / k / lgbt / lit / m / mlp / mu / n / news / o / out / p / po / pol / qa / r / r9k / s / s4s / sci / soc / sp / t / tg / toy / trash / trv / tv / u / v / vg / vp / vr / w / wg / wsg / wsr / x / y] [Home]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.
If a post contains personal/copyrighted/illegal content you can contact me at [email protected] with that post and thread number and it will be removed as soon as possible.
DMCA Content Takedown via dmca.com
All images are hosted on imgur.com, send takedown notices to them.
This is a 4chan archive - all of the content originated from them. If you need IP information for a Poster - you need to contact them. This website shows only archived content.