• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

PC Crashing

SirioAstarot

New Member
Joined
Mar 16, 2023
Messages
10 (0.01/day)
WTF!!!
I have been suffering from this problem for 20 days, sudden cuts, shutdowns and blocking, without any sense, I even burned a 750W power supply, I am going crazy!!
I don't know what the hell it is, I'm already tired of my computer problems.

CPU.png


Can somebody help me!
 
Can you be a little more specific? I get the gist of it, FX-8370E, RX560, 16GB DDR3, this means your computer's around 8 to 10 years old at this point in time. A computer of this age will more than likely exhibit several symptoms if maintenance hasn't been done regularly on it.

- You're probably due a complete disassembly, clean-up and reassembly of your machine
- Power supply may need to be replaced, if it was a bad quality one (you mentioned having "burned" it, I presume this is some language barrier but if it's bad, replace it)
- Depending on how long's it been on and the quality of your motherboard, it may have bad caps (rare, but can occur)
- CPU is more than likely overheating if you haven't replaced thermal paste in that long
- Similarly, the graphics card may require new thermal paste for the core and perhaps new thermal pads for the memory and VRMs
- Check all mechanical HDD storage for SMART errors
- Once all has been verified OK, consider reinstalling Windows

All in all your computer needs a complete maintenance pass. I'll be honest, though: don't put more than the bare minimum resources to get your computer working in order again. It's not worth spending any meaningful amount of money on this machine. Thermal paste, maybe a replacement heatsink, the power supply if necessary: but anything else, buy a new one. This PC is super slow and super out of date.
 
As Dr. Dro said above, can you elaborate a bit more on specifics? Do you notice any patterns? Are you doing a certain task when it tends to happen, etc? If you can narrow down a bit more we may be of more help.
 
As Dr. Dro said above, can you elaborate a bit more on specifics? Do you notice any patterns? Are you doing a certain task when it tends to happen, etc? If you can narrow down a bit more we may be of more help.
Hi,
It's random, it happens sporadically, there are no patterns like if I do this it happens.
I think it is due to the processor overheating.
 
Does the whole machine reboot? Black Screen, system acts like it is powering on? I would try a different power supply, I had this issue with my old(way back machine old) 8350 and a GTX 970. Turned it was the cheap ThermalTake 750w PSU. I used my brothers 900W for testing and it never crashed. I replaced it with the one I am still using today.
 
@SirioAstarot - please answer sLowEnd's question. It is critical we understand the problem if you expect us to give you sound and applicable advice.

I think it is due to the processor overheating.
80.5° is pretty warm but it is NOT hot or "over" heated.

While your symptoms could suggest a heat issue, they also could easily suggest a power issue too. Since EVERYTHING inside the computer case depends on good, clean, stable power, you need to ensure you are providing it. Swap in another, known good, PSU to see what happens.

Note too, the CPU is not the only heat sensitive device that can cause a system to crash. You might try removing the case side panel and blast a desk fan in there.

I would try one stick of RAM at a time too.
 
WTF!!!
I have been suffering from this problem for 20 days, sudden cuts, shutdowns and blocking, without any sense, I even burned a 750W power supply, I am going crazy!!
I don't know what the hell it is, I'm already tired of my computer problems.

View attachment 335156

Can somebody help me!
Hold up a sec - System specs PLEASE so we can know exactly what hardware you're dealing with and go from there.

You're talking about AM3+ (FX) here anyway and there are several things that can cause issues and it's aready been pointed out (Correctly) the parts do have some age on them by now anyway.

@Bill_Bright FX doesn't like temps much over 63c under load as a norm anyway, if it's really hitting 80c it's going to act up and I know that from all my time of dealing with FX chips/setups, they are VERY finicky about operating temps.
 
1708265618471.png


I think he may need to clean it like @Dr. Dro suggested.

35W and 80c is a little crazy..
 
35w and 80c??? that's not normal.....

Clean it and repaste the CPU, please
 
Sorry Dr. Dro and Jacobino III but repasting the CPU is not sound advice and sadly, is prematurely suggested WAY WAY too often, and unnecessarily.

There is no reason to replace the TIM (thermal interface material) AS LONG AS the cured bond between the mating surfaces is not broken. If the bond is still intact, the TIM is still doing it job of blocking insulating air from getting in between.

Note that TIM can easily last 10, 15 years or longer, again, AS LONG AS the cured bond is not broken. Remember, the purpose of TIM is to fill the microscopic pits and valleys in mating surfaces to force and keep insulating air out. Even if the TIM dries, the solids that are left behind are still occupying those pits and valleys.

Note the ONLY reason TIM comes in a liquified form is so it can easily be squeezed out of the tube or scooped out of a tub, then spread evenly and smoothly across the die.

Yes, replacing the TIM "may" provide a "few" degrees of additional cooling but the fact remains, if one "needs" those "few" degrees to prevent crossing over thermal protection thresholds, then they are already way too close to that threshold in the first place! There are more urgent cooling issues that need to be addressed first - like case cooling.

FTR - there is not a single CPU, GPU, cooler, motherboard, computer or even TIM maker that says or recommends users replace the TIM just because it is X number of years old. Not one!

"IF" there was no risk involved, my advice might be different. The problem is, there is the potential for significant risk of damage to the socket, mounting mechanism, and particularly the CPU from accidental mishandling and/or ESD. And for what, maybe 5° better cooling?

There is also the risk of common mistakes like failure to thoroughly clean the mating surfaces first. Or applying too much TIM.

If the cured bond is broken then absolutely the cooler must be removed, the mating surfaces thoroughly cleaned, and a "thin-as-possible" but thorough coverage, fresh, new layer of TIM be applied. Otherwise, leave it alone!
 
As I tend to not disagree with what you are saying @Bill_Bright, If the OP has never cleaned this PC and its running that hot at what appears to be Idle, I would recommend take the cooler off to clean it. (it probably needs it)
 
Are you overclocking? Have you performed RAM tests...
 
Sorry Dr. Dro and Jacobino III but repasting the CPU is not sound advice and sadly, is prematurely suggested WAY WAY too often, and unnecessarily.

There is no reason to replace the TIM (thermal interface material) AS LONG AS the cured bond between the mating surfaces is not broken. If the bond is still intact, the TIM is still doing it job of blocking insulating air from getting in between.

Note that TIM can easily last 10, 15 years or longer, again, AS LONG AS the cured bond is not broken. Remember, the purpose of TIM is to fill the microscopic pits and valleys in mating surfaces to force and keep insulating air out. Even if the TIM dries, the solids that are left behind are still occupying those pits and valleys.

Note the ONLY reason TIM comes in a liquified form is so it can easily be squeezed out of the tube or scooped out of a tub, then spread evenly and smoothly across the die.

Yes, replacing the TIM "may" provide a "few" degrees of additional cooling but the fact remains, if one "needs" those "few" degrees to prevent crossing over thermal protection thresholds, then they are already way too close to that threshold in the first place! There are more urgent cooling issues that need to be addressed first - like case cooling.

FTR - there is not a single CPU, GPU, cooler, motherboard, computer or even TIM maker that says or recommends users replace the TIM just because it is X number of years old. Not one!

"IF" there was no risk involved, my advice might be different. The problem is, there is the potential for significant risk of damage to the socket, mounting mechanism, and particularly the CPU from accidental mishandling and/or ESD. And for what, maybe 5° better cooling?

There is also the risk of common mistakes like failure to thoroughly clean the mating surfaces first. Or applying too much TIM.

If the cured bond is broken then absolutely the cooler must be removed, the mating surfaces thoroughly cleaned, and a "thin-as-possible" but thorough coverage, fresh, new layer of TIM be applied. Otherwise, leave it alone!

This is an AMD FX processor, and its Tjmax (junction temperature maximum) is 70°C, this processor is definitely overheating. FX likes to run cold, the same goes to the Phenom II.

Anyway... it doesn't hurt. They're getting >80 and the processor isn't even under full load as you can see on the OP. Thermal paste is cheap, and it takes 5 minutes to reapply. No reason why not do it, IMHO, especially if you're already cleaning the entire PC out. The vast majority of TIM are non-curing these days and I'm willing to bet that the one installed there is likely of this variant, which means at this point, after thousands of thermal cycles over the years, it's probably turned into sand or at the very least, leaked out (this also happens).
 
Are you overclocking? Have you performed RAM tests...
It's default clockspeed is only 3300MHz and his screenie shows it ranging anywhere from just over 1400 to a little above 4300MHz, so either he's doing it or it could be boosting up to 4300 while it runs.

CPU voltage is fine though even the max shown for it isn't that much and should be tolerable with ease even on air but the fact it seems to be spiking clockspeeds upwards so much that has to be part of it.
Yes, FX chips do not like being too hot at all as said earlier, they do respond well to voltage anyway so there is no need to volt it up for the clocks he's getting from it.

I also agree - If the cooler needs cleaning out a fresh re-TIM of it won't hurt a thing and as others have already pointed out, it probrably needs cleaning anyway.
 
I‘m not quite sure what else I can add here as it’s been suggested already here. What I would do first is try to get some fresh air to that system. Open the side panel, put a house fan blowing into it, make sure you can get the temps as cool as possible. Dust it, clean it out as good as you can. If you know how, re-apply your TIM even though a lot of times it is not the issue, but worth checking it out. You can test your RAM, if you have a good known working PSU, swap that one in, if all hardware tests fine, then try taking it out of the case? You have parts laying around you can try to swap out?

Many years ago I had I think it was a Phenom II system, and I had this weird issue kinda like what you were having. Random reboots, shut downs, etc. I went crazy, hardware tested fine, etc. I got fed up one day and took it out of the case, powered it up on a test bench, and the issue dissappeared. I came to the conclusion something in the case was causing a intermittent short of some kind. I never bothered to put it back into the same case, instead got a new case. Not saying this is your issue but I think if all else fails, it’s worth a shot.
 
As I tend to not disagree with what you are saying @Bill_Bright, If the OP has never cleaned this PC and its running that hot at what appears to be Idle,
Not sure what you disagree with here. Everything I said was true.

You can follow the link in the 1st line of my sig to see if I might have a little experience with hardware support and maintenance. I am not speaking as a hobbyist. But as a certified electronics tech with over 50 years working in the field as my profession. I have personally seen "OEM" TIM still adequately cool their mission critical devices after 20 years!

You can also see by the 4th line in my sig that I take heat (and thus cooling) seriously.

TIM does indeed last years - decades even. Yes, it degrades a little over time, but the typical gain from a new application is usually just a few, 5 - 6, degrees. So again, if you "need" that just to avoid crossing thermal protection thresholds, you are already running too close caused by other more urgent cooling needs that need to be addressed first.

Can you show us where Intel, AMD, NVIDIA, Gigabyte, ASUS, MSI or any TIM makers recommend regular changing of the TIM?

I would recommend take the cooler off to clean it. (it probably needs it)
I also agree - If the cooler needs cleaning out a fresh re-TIM of it won't hurt a thing

Sorry, but those comments make no sense at all! ABSOLUTELY, if the cooler needs cleaning, then ABSOLUTELY, clean it! And make sure the rest of the case interior is clean of heat-trapping dust and dirt too. But there is absolutely NO REASON the cooler needs to pulled to clean it! Take the computer outside and blast the cooler, fans and the rest of the case out of compressed air - preferably with a properly equipped/filtered air compressor. But if that's not available, cans of compressed dusting gas and a soft, natural-bristled dust/paint brush will do.

I say it again, AS LONG AS the cured bond is not broken, the TIM does not need to be replaced just because it is X years old. I will concede that I am assuming the TIM was properly applied in the first place, and that it has been doing its job effectively - as it appears to have done here.

As far as "a fresh re-TIM of it won't hurt a thing", I never said or even implied otherwise. What I said was, there is a significant risk of damage from "accidental mishandling and/or ESD". I also said there is the risk of mistakes due to improper cleaning of the mating surfaces or applying too much TIM.

So how does the cured bond get broken? Typically through abuse - dropping the computer on the floor or rough handling during transport. This is especially a problem with tall coolers in tower cases, and why I recommend they be removed for transport or shipping - particularly if the computer will be out of your possession/care and watchful eye.

Ironically, we have seen many times in the shop where the cured bond was broken by the user twisting the cooler to see if the cooler was loose! They ended up breaking the bond in the process! :rolleyes:

If the cured bond is broken, or you suspect it might be broken, then absolutely, pull the cooler, thoroughly clean the mating surfaces, properly apply a fresh, new, thin-as-possible" layer of TIM - while observing properly ESD precautions. But if the cured bond is not broken and you are crossing thermal protection thresholds, you have other, more urgent cooling needs to deal with first.

And if the cooler is caked with dust mixed with cat oily hair and cigarette smoke residue, you still don't have to remove the cooler (thus breaking the bond) to clean it.
 
I said I dont disagree.. replacing the paste at regular basis is not needed. But, if you have never cleaned, or its been an extended period of time from the last time you cleaned the cooler, pulling the cooler off and deep cleaning it maybe what is needed. His idle temps are excessively high, so I would recommend pulling the cooler and cleaning everything on it and repasting it. Thats all I was saying..
 
I said I dont disagree..
:oops: Doh! Indeed you did. It's those tiny 3 letter words my mind too often skips right over that get me every time.

My apologies. I suffer from chronic GCF* with occasional acute flare ups. I'll blame it on that.

*GCF - geriatric cranial flatulence. ;)
 
that poor cpu getting way too hot. reseat/repaste the cpu-cooler and reset everything back to default in bios. update bios to the latest version. then start again with ur testings. after that id carefully start to undervolt it. i have an FX8350 machine with a gigabyte board and 32Gb ram, i am playing a lot with it just for fun
Screenshot_1.png
 
Gentlemen, do not fight, there is an issue that is public knowledge, the Vishera architecture has high temperature problems, they were the last models before the ZEN came onto the market, back in 2014, yes, my PC is very old and it is I need to change it, although I am buying the components little by little.

In cooling I have a Thermaltake Ux100 ARGB 120mm AM4, I put liquid metal thermal paste on it again, the case is Thermaltake Level 20 RS RGB with two 20cm front fans and I opened the side cover, everything lowered to 50° C.

Regarding the power supply, I had a Gigabyte GP-P750GM 750w, since it burned I sent it to change it, it covers the warranty, while I bought another one, a 600W Thermaltake Smart White TR-600.

I entered the BIOS trying to perform Underclocking and Undervolting, but it wouldn't let me, everything was on automatic. :(

This is already something that exceeds my knowledge and I am in a situation beyond my reach, I live in Argentina and we are suffering the most intense heat wave in the last 20 years, with thermal sensations of 50°, a true hell.

Now everything seems normal, I will let you know if it fails again, although I don't think it was due to Electrostatic Discharge or small cuts, it was all the damn temperature, I'm not going to add anything else, not even liquid cooling or anything else, I have everything new, the only old thing is my Gigabyte GA-990FX Gaming v1.1 motherboard and the AMD FX-8370e processor, the next thing will be Zen 5 on Chipset X870E with RDNA 4.

@Bill_Bright @Dr. Dro
Thank you all very much for the suggestions and recommendations, they have helped me a lot.
 
In cooling I have a Thermaltake Ux100 ARGB 120mm AM4, I put liquid metal thermal paste on it again,
Just make sure that you don't spill it anywhere on the motherboard or other components
 
In cooling I have a Thermaltake Ux100 ARGB 120mm AM4, I put liquid metal thermal paste on it again, the case is Thermaltake Level 20 RS RGB with two 20cm front fans and I opened the side cover, everything lowered to 50° C.
BRUH NO

DO NOT USE LIQUID METAL ON AN ALUMINUM HEATSINK LIKE THAT ONE

Liquid metal will corrode the heatsink very quickly
 
BRUH NO

DO NOT USE LIQUID METAL ON AN ALUMINUM HEATSINK LIKE THAT ONE

Liquid metal will corrode the heatsink very quickly

Oof that is a big no-no indeed, by now the mess is probably already made... Liquid metal should be used exclusively with copper heatsinks
 
Can you be a little more specific? I get the gist of it, FX-8370E, RX560, 16GB DDR3, this means your computer's around 8 to 10 years old at this point in time. A computer of this age will more than likely exhibit several symptoms if maintenance hasn't been done regularly on it.

- You're probably due a complete disassembly, clean-up and reassembly of your machine
- Power supply may need to be replaced, if it was a bad quality one (you mentioned having "burned" it, I presume this is some language barrier but if it's bad, replace it)
- Depending on how long's it been on and the quality of your motherboard, it may have bad caps (rare, but can occur)
- CPU is more than likely overheating if you haven't replaced thermal paste in that long
- Similarly, the graphics card may require new thermal paste for the core and perhaps new thermal pads for the memory and VRMs
- Check all mechanical HDD storage for SMART errors
- Once all has been verified OK, consider reinstalling Windows

All in all your computer needs a complete maintenance pass. I'll be honest, though: don't put more than the bare minimum resources to get your computer working in order again. It's not worth spending any meaningful amount of money on this machine. Thermal paste, maybe a replacement heatsink, the power supply if necessary: but anything else, buy a new one. This PC is super slow and super out of date.
the problem is FX CPU...:roll:

WTF!!!
I have been suffering from this problem for 20 days, sudden cuts, shutdowns and blocking, without any sense, I even burned a 750W power supply, I am going crazy!!
I don't know what the hell it is, I'm already tired of my computer problems.

View attachment 335156

Can somebody help me!
just change this BS FX CPU base unless you wanna burn not only PSU but your house.....
 
Back
Top