I fucked up. Long story short, I pulled cpu cooler off of my motherboard, but the processor held on to it. What I want to do is have the processor and heatsink back on the board, but the heatsink is preventing me from having a good enough view to line up the pins correctly. What do I do, /g/? I tried pulling the processor off the cooler, its on too tight, and I tried using floss as well to no avail. I apologize for the terrible pic and if this issue belongs somewhere else.
>>47445367
what the fuck kind of shit thermal paste have you been using anon
2 birds 1 stone. Nice.
Pretty much just heat it up a tiny bit and slide it off instead of prying.
Then redo the thermal compound.
If it's thermal adhesive though your fucked have to use chemicals and they generally will melt the processor.
Also it looks like a vishera FX series.
Use rubbing alcohol to dissolve the paste and separate the CPU.
>>47445367
remount it into the PC, and run a CPU stress test. the heat might help loosen the paste.
>>47445367
Just go to fucking home depot or ace hardware and get a plastic putty knife and be gentle as fuck scraping it off so it doesn't fly across the room.
>>47445367
1. Remove the processor from the heat sink. Don't fuck up the pins.
2. Re-seat the processor
3. Acquire thermal paste
4. Disregard heatsink
>>47445404
Btw no direct heat on the processor.
>>47445367
>the heatsink is preventing me from having a good enough view to line up the pins correctly
how can you even have this problem
the socket is keyed
you can't put it in wrong
>thats_what_she_said.xpm
Jesus what. Why didn't you twist it off or turn on the pc to heat it up first
Use a hammer and delid it, you can have better temp.
>>47445423
no I know that stock cooler. OP needs a BULLDOZER
>>47445433
hes afraid of bending them
>>47445367
what kind of processor socket do you have OP?
>>47445478
I believe it is an AM3+ socket
>>47445516
Found the issue.
>>47445423
I feel like my safest option is to wait until tomorrow so I can get a clean razor or knife like in your pic to scrape it off
>>47445367
This happens all the time with AMD's sockets when thermal compound dries and terms to cement.
You just have to keep trying to twist it until it breaks free. Do not scrape at it with anything. Don't use any sort of penetrating oil. Don't try to grab the CPU package with pliers.
>>47445540
>waiting until tomorrow
maximum pleb
>>47445404
Its a Phenom II. How do I safely apply heat? Sounds like it could possibly fuck something up.
>>47445540
Buy a disposable scalpel. Trust me.
>>47445367
OK here's what you do. You stick it in the microwave face up and set it to defrost for two minutes. The defrost setting will keep it from frying the processor and the heat will loosen the thermal paste. Once it's done pry it apart. ??? PROFIT!!!
>>47445554
My CPU likes it when I grab his package.
>>47445554
>Don't try to grab the CPU package with pliers
maximum kek imagining someone trying this.
>>47445565
anything that can't generate heat over 60C should be fine. try a hair dryer on high blowing through the fins for 10 minutes or so.
>>47445565
Use a hairdryer.
>>47445367
Get a piece of floss, wedge it in, and saw it out. Heating it up w/ a hair dryer will help a bit too.
>>47445593
It happened in a thread on /g/ not too long ago.
There was also a guy here who bent a ton of pins trying to remove his stuck CPU.
>>47445554
>twist
>doesn't come loose
>twist harder
>still doesn't come loose
>"just keep twisting, anon"
>maximum twist achieved
>cpu flies across room
Why does this always seem to be an issue with AMD and not Intel? Is this the reason why Intel makes their IHS convex?
OP here, so is my safest route here to loosen the paste with a hair dryer, then try to twist it off?
>>47445767
Its just the locking mechanism on the socket itself. Intel's sockets use a big metal brace to hold the CPU down against the pins. AMD's sockets have no such mechanism.
>>47445821
>AMD's sockets have no such mechanism.
Interesting, I didn't know that.
>>47445821
Don't they still lock by holding the pins or is it just too weak?
>>47445875
it's a gentle grip, but strong enough that doing what OP did could risk bending all the pins at once.
>>47445902
Seems like a dumb system. Does Intel have a strong patent on their socket design or something?
>>47445875
Its just light tension on the pins. If the socket held on to them any tighter you'd be lightly to rip a pin off of the CPU instead of just pulling the CPU out of the socket.
>>47445930
>lightly
Likely. Fuck.
>>47445924
I'm wondering this as well. Home come AMD isn't using LGA or some other kind of pinless socket?
>>47446056
Because AMD is bargain bin material.
>>47445924
>>47446056
It looks like AMD had a pinless opteron back in the P4 era. I can't find anything on why they abandoned the design.
>>47445924
Intel didn't create the locking mechanism they use. AMD didn't create theirs either.
>>47446056
too practical for autistic people
Your best option is to fap until you cum on your CPU. Cum contains enzymes that rapidly break down thermal paste so the CPU will be free in seconds without any prying
Happens everytime with my FX8350.
You can try this.
>>47445406
or
>>47445612
stop having baby grips you sissy faggot.
>>47445924
Yeah, it is really dumb, you can even pull the CPU without disabling the lock, it happens all the times when trying to remove the cooler.
First time is always scary as fuck, try to picture it, you are trying to pull off the cooler like you've done with a tons of cpus before but this times shit comes together with the cpu stick on it, the first thought that comes to you mind is "fuck"
>>47446160
If you spend enough time on /g/ then you too can start cumming pure rubbing alcohol.
>>47445367
Shiet nigga, I feel you.
Holt it at an angle, and drip some Articlean 1 down the side. You can try goo gone, but I've heard that it'll ironically leave goo behind when it dries. Rubbing alcohol doesn't really work since it doesn't really break down the thermal paste. And try to slide it, especially into some Articlean 1 or rubbing alcohol. Don't pry.
>>47446056
LGA doesn't have any benefits other than a more easily damaged socket.
The pins on a PGA-ZIF socket are really fucking hard to bend or even snap off unless you try really hard to fuck it up. You can't bend a whole row of pins with just your fingers, you have to bend them one or two at a time.
Fuck, i've even butterfingers'd an Athlon II and dropped it on the floor and it was fine. One or two bent pins, nudged them back into place with a needle and it ran like a champ.
I could maybe see problems in the future when they have to have >1K pins on a chip, but for now they have a smaller socket with a comparable amount of pins on it vs Intel's LGA setup, and it's tougher to boot.
>>47445423
This
Something plastic won't damage it, as long as you don't slip or send it flying
You can also try rotating it off by putting pressure on the corners, but be careful not to slip and press on the pins
>>47446419
>lga
>no benefit
Are you forgetting how much cpus cost relative to motherboards?
>>47446562
What benefit does LGA have then?
You can't damage a PGA socket unless you really ream on it with a hammer.
LGA can be damaged just by dropping the fucking CPU itself into it wrong.
PGA pins, as I fucking pointed out, are much harder to bend, and are easily user-serviceable as long as you aren't michael j fox or a retard.
are you really this fucking retarded?
>>47446607
>What benefit does LGA have then?
Pin density.
Do you know how fucking huge a 2000+ pin PGA socket would be?
>>47446106
they still do LGA for opterons. i think they stuck with PGA for desktop for compatibility reasons. the number of pins has stayed almost the same since socket 939 in the early 2000s.
it was a neat idea at the time, but it's kinda stupid now. i hope K12/Zen are LGA
