Paste

>>59046151
Sounds like am add your thing

>>59046151
Good to know.

>>59046151
dump the good paste and revert bad to el cheapo. see if it reverts bad to 2.7GHz. Your improvement is probably just down to having fresh paste instead of some old crusty ass shit.

>>59046441
Only posted this because many anons told me the cooling in my laptop was trash because it's a laptop and good paste wouldn't make it much better, if at all.

I'm amazed at the difference to be honest.

>>59046473
I'm sure. at the same 2.7GHz there's a roughly 15°c difference in temperature so the paste has definitely affected the temperature.
You can manually adjust cpu frequency for most processors with throttlestop.

>>59046566
It's pretty good advice. What model laptop do you have?

File: mobo.jpg (1MB, 4047x2621px) Image search: [Google]

1MB, 4047x2621px

>>59046588
It's a Sony Vaio.
VPCSA28GG.

To be fair though, the cooling is horribly insufficient for the 60w tdp of this laptop, but for just the CPU alone it's plenty.

The temps I provided were under a torture test, not standard usage temps.

>>59046814
Will also test on my desktop at a later time to verify how much better the good paste is, since I'm wondering how this being a laptop affected the results of the paste.

Anyone got a list of decent pastes? I'm due to replace CPU and GPU paste after 5 years.

File: eeeeeee.png (34KB, 425x487px) Image search: [Google]

34KB, 425x487px

sony VPCEB3S1E here
should i do the paste replace too?

>>59046814
>60w tdp
>laptop
And I thought my T420 with a 45W tdp quadcore was a lot.

File: mobo.jpg (2MB, 3075x3103px) Image search: [Google]

2MB, 3075x3103px

>>59047092
35w i7 2620m and 25w HD6630m, they're connected to the same heatsink m8.

>>59047071
Yes. Just a heads up, my Vaio is a piece of shit on the inside so yours might be difficult to deal with too. Well worth it for the temp gain though. The chink paste reduced temps a bit compared to stock when applied, but the Mastergel Maker was significantly better.

Have a look at what's in stock at your local stores and which is best for the price. As an ausfag my options are quite limited so I'm happy with what I was able to get.

>>59047022
gelid gc extreme is the best non retard priced paste, also I think second best paced overall before you go into conductive liquid metal/solder pastes

File: 1457023943558.png (15KB, 385x339px) Image search: [Google]

15KB, 385x339px

>friends' kids' gpu fan dies
>i go to replace it with a fan from a dead gpu
>replacement too big for shroud, but can't remove shroud without taking the heatsink off
>no paste
>whatever, replace old dry paste with minty-fresh toothpaste
>works fine

>>59046151
> >temp doesn't change much (within margin of error)
Did you check it at full load?

File: after .jpg (98KB, 425x487px) Image search: [Google]

98KB, 425x487px

>>59048386
it was really easy, just remove all the screws from the bottom. removing the hard paste was the hardest part

>>59048469
Yep.

>>59048477
Lucky. Mine was all compacted and tucked into places it shouldn't fit to make the laptop smaller so pulling out the motherboard was a pretty big nuisance.

>>59048469
You're missing the rest. Temp didn't change but his processor wasn't throttling to maintain that temp.

>miracle gains from better paste
nah

>>59049006
how is yours at speed? i want to put a ssd in mine but a 500gb ssd is really expensive.
maybe max out the ram to 8gb as well but i never get over 80% ram use anyway

>>59049115
Mines got 4 SSDs in raid 0, gets nearly 1TB/s in read and write speed last time I checked.
It's pretty good.

>>59049072
>miracle gains
Lol wut

>>59046814
That heatsink is TINY

>>59048396
>gelid gc extreme
Just checked this one and am put off by a bad review where it says the paste doesn't last long. I'd rather have long term usefulness than temporary bleeding edge performance. What thermal paste does that?

>>59046473
>Your improvement is probably just down to having fresh paste instead of some old crusty ass shit.
That's also a reason to use good paste instead of no-name zinc-oxide crap: the good paste takes like a decade to dry out and need reapplication, instead of what, a year or three?

>>59046566
yes but the point is, if you're bought a fresh tube of el cheapo and replaced the paste with that, you'd probably see a big improvement just because new paste works much better than years-old dried up paste, regardless of brand.

basically to apply proper rigorous testing to this you should test a new application of the old stuff and see what the temps are like, that would tell you if the improvement is due to the paste being fresh or due to it being a better brand.

good/bad paste does make a difference, but in almost all tests there's a max delta of about 5 degrees between bargain stuff and the super expensive stuff. proper application and replacing it when it gets old is like 90% of the difference in temps.

>>59053312
Just look for the thickest paste you can find.

>>59053388
I don't know shit about computers. What about Arctic Silver 5? That seems to be the consensus best.

>>59053368
It could have also been poor application of paste? I doubt it would have been dry though since he still had the tube and it looked fairly new.

>>59046151
I'm using Arctic Silver 5. My laptops are significantly cooler.

>>59046151
link?