[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/wsr/ - Worksafe Requests


Thread archived.
You cannot reply anymore.




File: space heater.png (43 KB, 609x99)
43 KB
43 KB PNG
Ive been using my PC for a few years now but with games becoming more and more demanding I noticed more frequent crashes and very high temps form my cpu. Its a ryzen 2600x with a vega 56.

My ryzen sometimes goes all the way up to 96°C and is around 60 to high 60s while idling and just using firefox and discord.

The PC also crashes randomly when I do stuff that is demanding and its not a usual crash the entire pc just instantly turns off (i.e. no blue screen or anything) and i have to turn off the powersupply for a few seconds until i can turn it back on.

The cpu cooler should be good enough but I think it takes the hot air from the gpu to cool it which might cause the high temps.

Are the temps alright and the crashes are due to something else like the powersupply or should i turn the cpu cooler to blow in a different direction?
>>
>>1263261
>is my CPU overheating?
>96°C
Yes.
>The cpu cooler should be good enough
If you've had it "a few years now", it will be full of dust and grease. Take the fan off and clean it, or if that's too scary, poke a q-tip through the fan (when it's off!) and clean the crap off that way.
>>
>>1263281
Thanks, usually I should just be able to remove the 4 screws to take the cooler off right?
Ive built a pc before but not this one so I want to be as safe as possible.
I also guess that I need to clean the thermal paste off both the cpu and cooler to reapply it.
>>
>>1263292
>I should just be able to remove the 4 screws to take the cooler off right?
If at all possible, see if you can just take the fan off the cooler. It's a known defect of the AM4 socket that the cooler can glue itself to the CPU and rip it out the socket or even pull the pins or heatspreader off.

Mitigations include getting the CPU as hot as possible first, being gentle with the cooler, and using a spudger or a dental floss "cheesewire" to remove the heatsink from the CPU while it's still in the socket.
>>
>>1263292
>I also guess that I need to clean the thermal paste off both the cpu and cooler to reapply it.
Yes. I usually get the bulk of it with a dry paper towel, then finish off with isopropanol on a q-tip.
>>
>>1263300
Damn thank you I gotta be more careful than I thought. Honestly I can hopefully just attach the fan to the other side of the heatsink to get it further away from the GPUs hot air.
>>
Just to add, it doesn't often hit 96C while playing Monster Hunter Rise (not for long tho) it went into the 80s, does this still count as it overheating?
>>
>>1263323
Even 80 is not a fantastic temperature, are you sure it's not full of dust? Did it do that when it was new?
>>
>>1263328
On the contrary, I'm pretty sure it is full of dust I'm just trying to find any other reason without having to take anything out of the PC because I really don't want to break it.
Didn't really check the temps when it was new but pretty sure it did not reach these temps.
>>
>>1263308
>I can hopefully just attach the fan to the other side of the heatsink
What you want to be doing is coming up with a plan for how the air is going to go. It's definitely possible to get stuck in the weeds on this, you can get perfectly good results by picking two directions (up and back are typical), and having every fan blow in one or the other. So if you picked "up" and "back", no fan can blow "down" or "frontward". That way you don't have your fans working against each other.

There's no problem with the CPU ingesting the GPU's exhaust, so long as the air's all going in the same direction. It's not ideal, but the air will still cool the CPU, and if the CPU starts making more demand for air, cold air will flow past the GPU.

On that note, make sure you haven't got the GPU exactly level with a drive cage, forming a makeshift divider that air can't get past.
>>
>>1263300
>Mitigations include getting the CPU as hot as possible first,
I have found that simply turning the heatsink relative to the CPU a bit works wonders. Never had problems with that technique.

>>1263329
Get some compressed air and blow the fuck out of your case. Outside, for heaven's sake.
>>
>>1263332
Thanks for the advice, I have the intake for the air in the front and the exhaust at the back but gotta test what direction the GPU and CPU fans blow. I mean they are facing eachother (as in, the logo of both fans face one another) so its probably not going in the same direction.
>>
>>1263335
>Never had problems with that technique.
If you do it too hard, you'll pull the pins off. I've never had a problem doing that on an LGA processor though, so I assume you'd have to be pretty simian to twist a CPU out its socket. I don't imagine there's that much potential for fuckup so long as you stop pulling if it's not coming.
>>
>>1263340
I don't think you could do that if you wanted to. You are twisting against a LOT of surface area at once -- all the sideways strength of the 1200 pins together adds up to something you are not capable of breaking like this.
>>
>>1263371
You only have to break one though.
I'm not about to go test this or anything, but the general principle is that you're not fighting them all at once, there's going to be an outlier that takes most of the force. Then once it goes the next one takes most of the force, and so on. There's literally nothing holding an AM4 CPU in the socket other than its pins, so while it will probably go okay, you don't want to be taking unnecessary risks just to save five minutes. Go run prime95 and get some heat in the paste. Stop at the first sign of trouble. If it's not going, cheesewire it. And so on.

I mean I don't think it's likely doing that will cause trouble, and I'd have no problem doing it on 1151, 1200, etc., but I'd be reticent about trying it on AM4.

https://www.youtube.com/watch?v=kMox-tOmgJw
>>
>>1263337
Wait I'm retarded they obviously do not face eachother.
>>
>>1263261
>idling
>96c
yes. you likely need to repaste your cpu also try switching to power saving mode. I'm using a 5600X with a small L9a cooler and on powersaving mode and rn with a bunch of tabs open and playing MCC it barely goes above 55c
>>
>>1263404
OP said 96 was heavy load and idling was 60

You're not wrong that 96 is high for load, but it's not the hellscape you're imagining.
>>
>>1263261
>is my CPU overheating?
Open up the side panel and see if dust is clogging or fine grit is coating the heat sink fins.
>>
>>1263261
>The PC also crashes randomly when I do stuff that is demanding and its not a usual crash the entire pc just instantly turns off (i.e. no blue screen or anything) and i have to turn off the powersupply for a few seconds until i can turn it back on.

That sounds like thermal protection shutdown. To prevent more over temp it instantly cuts off power so you stop frying the die even worse than you already are. It won't turn back on until it cools enough.

Try using something like Open Hardware Monitor to log the cpu temp, compare the temp graph vs the BIOS settings because you can have your mobo make nice beeping sounds when the cpu gets too hot.


>>1263323
>went into the 80s, does this still count as it overheating?

According to amd: maybe. 95 centigrade is apparently ryzen's shut off temp. It can become unstable over 65 for overclocking.

>>1263329
>Didn't really check the temps when it was new but pretty sure it did not reach these temps.

If it had reached those temps it would have been thermally shutting down.
>>
>>1263529
>it instantly cuts off power so you stop frying the die even worse than you already are
That's ridiculous, think about it for a second: how would a CPU be able to correctly drive the ATX PSU signal lines (which are not a standard) on every motherboard ever, including motherboards that are newer than it is?

It may well be the computer is having heat-related problems, but the mechanism you're describing doesn't exist, or this video would have a different ending: https://www.youtube.com/watch?v=my7BJRQ1OqM
>>
>>1263586
>doesn't know something exists, says it can't exist

Your choice to remain in ignorance.
>>
>>1263586
>ATX PSU signal lines (which are not a standard)
They are. Motherboards turn PSUs on or off with a control line, that is how the power button has worked since the first ATX motherboards. (AT power supplies had a physical on/off switch, which is why windows 95 had the "it is now safe to turn off your computer" thing, but that was many decades ago).



Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

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.