Thermal paste
Images are sometimes not shown due to bandwidth/network limitations. Refreshing the page usually helps.
You are currently reading a thread in /g/ - Technology
You are currently reading a thread in /g/ - Technology
How much of a difference is there between cheap Chinese thermal paste and something decent like noctua or arctic cooling?
>30g for $1.40
Enough to justify the cost of the higher quality stuff for your personal PC, not enough to justify the higher cost when building other people's without charging them an upgrade fee.
Just buy Arctic Silver 5, you fucking idiot. Every other TIM is complete garbage.
It's $5-7 bucks.
I only use the cheap Chinese shit (that I order in bulk) to fix things like Xbox 360s and other people's computers.
>>51236011
>Arctic Silver
That stuff has been outclassed for years. Literal dog shit.
>>51236011
How much difference is it?
This is 30g of paste, pretty good value.
If it's less than 3 or 4 degrees difference i won't bother, since it's just so much cheaper.
>>51235988
This is pretty much what it comes down to.
Which one is better is highly opinionated, as pretty much all top rated thermal pastes are identical, within a 1-2C variance, when it comes to temps, so just grab whichever one is on sale at the time.
>>51236230
doubt it's even 3-4 C.
you'll get bigger diffirences in any tests from the way you spread it, room temperature or even cpu load than from paste itself.
Even as a guy who works in a computer repair shop, where my job is to jew people out of every shekel they have, there is no reason to not use high quality thermal paste. Its not that big of an expense.
Ive put TX4 on everything from a 14 thousand dollar mac pro, to some guys 1998 MicronPC tower he wanted cleaned.
Its not a big expense for the large jump in quality and lifespan.
>>51236021
>he's afraid to use conductive paste
>I-it's outclassed f-faggot
I'm happy with Prolimatech PK-2. Non-conductive, performs as well as AS5 and is not that viscous to make application hard. Replaced my GPU's paste after 4 years and saw a 6 degree decrease in load temps as well as 4% less fan speed.
>>51236293
But I'll assume I'll have to replace this more often?
I don't mind, 30g would be something like 50-60 applications.
I use that OP. It works fine.
>>51236421
not really.
>>51236470
>>51236453
Thanks guys, I'll just order this.
Can't really fuck up for only $1.40
Right, so let's use science to explain instead of random /g/ feelings.
Silver is one of the most thermally conductive materials that exists. Here's a rough list for you, they're in WmK (Watts per meter per Kelvin), which probably doesn't make sense to you so just understand that more is better.
>Zinc Oxide: 25
>Boron Nitride: 100
>Aluminum Nitride: 175
>Pure Aluminum: 200
>Pure Silver: 400
>Pure Diamonds: 2000
Those are the most likely things to be used for heat conductivity in your thermal paste. Everything else is either worse, retardedly expensive, or far off into theoretical land. Now, thermal paste obviously isn't composed of pure forms of any of those substances. They're all suspended in a grease-like matrix of other shit, which lowers the conductivity a lot. But you should still be able to see the difference between Zinc Oxide (what your Chinese shit is most likely based on) and Silver, which is what Arctic Silver (no shit) is based on.
>>51236747
Where does ceramic-based paste fall into this? What about natural peanut butter (80s staple, before mass hydrogenation)?
>>51236800
Ceramics are actually really good at conducting heat, but I'm lazy and couldn't find an average in the first two links I clicked on Google. Peanut butter probably falls somewhere around the "Might as well just use your own semen" tier.
Use cum.
>>51236747
>silver thermal grease
Yeah, I'm going to guess they mean silver colored, at this price...
>>51236860
that's a joke right? Why the fuck isn't everyone using chocolate?
>>51236888
lower is better anon.
>>51236888
>that's a joke right?
Of course not, go ahead and use chocolate.
>>51236860
>32 degrees
Why are they testing idle temps? The results would diverge if they ran a CPU stress test
>>51236888
...Because it's the worst one on the list? It shows right there it gets the highest temps, are you blind?
In other news, Mayonnaise is literally a better thermal paste than 40% of ACTUAL MARKETED THERMAL PASTES on this list.
>>51236860
>no mention of Gelid
Old as fuck.
>>51236882
>such as CPU, NVIDIA...
>"high insulation" the opposite of anything you want out of thermal paste
>"hot chips", not shitting on that, the phase "hot chips" just made me lol
>>51236993
>hot chips
Not sure how well this sauce tastes, not sure if I want it for my hot chips.
>>51237096
Hot chips is also slang for hot Cheetos (a spicy version of some "cheesy" potato crisps) where I'm from
