Overclock.net banner

1 - 18 of 18 Posts

·
Registered
Joined
·
103 Posts
Discussion Starter #1
Hi, back around Thanksgiving I upgraded my rig (Antec 900 with all fans active) with a new cpu & gpu.
AMD FX 8350 & ASUS strix gtx970. Plays everything I could ever want smooth as can be.

This past week something strange happened while I was playing R6:Siege online. In a multiplayer match my computer completely shut down suddenly. It just happened again today.

In the past I've been working on a laptop during the day and using my rig only for gaming, but the past week I've had the rig on all day for work and then started gaming. Based on the ASUS GPU monitor, nothing over 60 fahrenheit, but when I downloaded the CPUID HWMonitor I got readings averaging 100 F for the CPU. See image attached.


I wonder if with the warmer days from springtime my cpu is finally feeling the heat. One steam R6: Siege thread suggested the PSU or power strip might be to blame for sudden shut downs. MY Corsair TX750 is almost 5 years old come June.

I suppose now I'm mostly worried about the CPU overheating. The stock heatsink CPU was always running loud from day one - I just figured that was normal. Really wish I'd thought to dl a temperature monitor back then.

Any commentary on CPU or PSU would be greatly appreciated.
Could a fancier heatsink help? I don't want to water cool.
I'm also considering getting a new power strip, maybe a rack mounted one if it will be better protection for the summer's inevitable thunderstorms.
 

·
What should be here ?
Joined
·
5,735 Posts
I'd first try out a spare PSU. If your system goes down without throwing any errors, it can be your PSU. Also, what are your complete system specs ? Any other applications where it does this ?
 
  • Rep+
Reactions: RepCom1207

·
Senior OCN Member
Joined
·
39,857 Posts
  • Rep+
Reactions: RepCom1207

·
Registered
Joined
·
103 Posts
Discussion Starter #5
nah it's not that one. mine has yellow and white letters not orange. Its the TX750 V2 Enthusiast Series.
Today after being shut off all night I turned on my rig and checked the CPU temperature.
Should I replace the stock heatsink / fan and find a better one?
 

·
I Am Gooble Gobble
Joined
·
6,669 Posts
it seems to me by that screenshot that the cooler isnt seated properly.check the locking pins for it, if its not that then if you have thermal paste laying about, shut it down, clean cpu and heatsink off and re apply thermal paste
 

·
Master of Black Snow
Joined
·
19,338 Posts
Quote:
Originally Posted by RepCom1207 View Post

nah it's not that one. mine has yellow and white letters not orange. Its the TX750 V2 Enthusiast Series.
Today after being shut off all night I turned on my rig and checked the CPU temperature.
Should I replace the stock heatsink / fan and find a better one?
I would. The stock cooler for that chip while decent enough for browsing and fB gaming, it's not worthy of a gaming rig setup. You can get a reasonable CM Hyper 21* series cooler for a decent price.

I would also point out that likely the problem is that CPU being a heat monster. I've RMA'ed 2 FX CPUS due to heat issues causing heat dumps. I use [email protected] to stress and if anything will show a bad CPU in a hurry, that will. I had one dump not long after POST and one dumped within an hour after starting Stress. So likely you have issues at the CPU and your PSU is fine. So try for a Cooler Master Hyper HS first. Stress the CPU after installing the cooler and that should show you whether to open the RMA with AMD.
thinking.gif


http://www.newegg.com/Product/Productcompare.aspx?Submit=ENE&N=100008000%2050001333&IsNodeId=1&bop=And&CompareItemList=574%7C9SIA1N83YW6755%5E9SIA1N83YW6755%2C35-103-099%5E35-103-099&percm=9SIA1N83YW6755%3A%24%24%24%24%24%24%24%3B35-103-099%3A%24%24%24%24%24%24%24

I thought about it before submitting this post, if you have the watercooling heatsink, that should've been fine for cooling. If you have the typical AMD heatsink that comes with low end CPUs, that is the one that sucks for FX chips. You can mod them to fix the issue by cleaning the cheap TIM off and lapping the base using 1000, 1400 and 2000 grit papers. I have done a few(I was bored) and it actually made temps better than stock. So if you want to try it and see if it makes a difference, it's cheap enough to do and you won't be out $30 for the new cooler yet.
thumb.gif


~Ceadder
smil3dbd4e4c2e742.gif
 
  • Rep+
Reactions: RepCom1207

·
Premium Member
Joined
·
17,110 Posts
Instant shut down is almost always a faulty PSU, i've honestly never had one case where it was temp related on modern hardware. They all throttle and/or BSOD before heat becomes an issue.

There is absolutely no way it's 60F, that's below ambient temps which is impossible to do on air let alone a stock cooler, so the ASUS program is wrong. Also, no one uses F when talking about hardware temps, it'll only lead to confusion. 47C isn't close to the max temp that processor is rated at - I assume they are idle temps. Either way, stock cooler is designed for desktop load and sudden quick CPU spikes, not long gaming sessions. I still don't think your stock cooler is seated right though because your core difference is 13C.

I would replace the PSU and get a proper CPU cooler. Don't worry about what's causing it because you should do both.
 

·
Registered
Joined
·
103 Posts
Discussion Starter #9
Ah I'm an idiot. My GPU is idling at 39C.
Yesterday I tried a new monitor for the CPU and got 50C in R6:Siege main menu; 70C mid-match; then 30C after shutting off the game.
I'm definitely going to get a new heatsink to be on the safe side. This one was recommended by another forum member when I mentioned looking to avoid water cooling and keep the price under $50.
I was thinking after I get that hooked up I'll run the game again and see how the PSU holds up. But if you think the PSU is at the end of its days, then do you have any suggestions for reliable models?
 

·
What should be here ?
Joined
·
5,735 Posts
You could get a EVGA G2 650W for ~60$ but thats a best performance /$ PSU I would recommend. If you come across another one, do tell us and we'll let you know
 

·
Master of Black Snow
Joined
·
19,338 Posts
Hyper 212 is $30 from Newegg. But that one isn't too bad either.
thumb.gif


~Ceadder
smil3dbd4e4c2e742.gif
 

·
Registered
Joined
·
103 Posts
Discussion Starter #12
I just hooked up my brother's TX750 which is a later generation - his has 0 rpm tech so the fan doesn't spin until needed. I loaded up R6: Siege and hit 60C on the CPU in matchmaking. Lasted 4 minutes into the first round before everything froze and my keyboard became non-responsive.
So I guess that means my PSU was fine. It would seem my CPU getting too hot caused the abrupt system shutdowns.
I have just rummaged through the garage and found a stationary fan to set by case in place of the door.

For now the CPU is reading an avg of 40C while I'm on Opera. Going to boot up Killing Floor 2 to see how the CPU handles that game's more relaxed system reqs.
 

·
Registered
Joined
·
103 Posts
Discussion Starter #13
Well having the fan on meant I was able to play a full match of Siege and then shut off the game with no troubles. Temperatures were still crazy though.
^^ in game.

^^ continuing on in game, yikes...

So now I wonder is replacing the heatsink going to be enough? The CPU temps on the Mobo are higher than the CPU itself. Should I just bite the bullet and get a new motherboard? In the past I didn't want to spend much over $100 on the mobo but if it means quality and longevity with this hothead CPU, I'd be willing to go closer to $175 or whatever would be the best value for my needs.
 

·
Registered
Joined
·
103 Posts
Discussion Starter #15
if i get a better cooler does that mean I won't have to underclock it? or would that still be recommended for my ancient mobo
 

·
Registered
Joined
·
518 Posts
I don't think undervolting will be neccessary. I never tried the old stock cooler myself because of it's terrible reputation but from what I heard/read the cooling is insufficiant even for stock clocks/voltage. That would be in line with your expierince. I'm not entirely certain but I think you can even oc a bit (4.2-4.3GHz) with a Hyper Evo 212 or a comparable cooler.

Setting the voltage manually is always a good thing though, if you are willing to put in the time to find the lowest voltage needed for being stable with the clock you want to run. Everything on "Auto" will often set a bit higher VCore than actually needed.

Also, VERY IMPORTANT to temperatures is the airflow inside the case. User @doyll has an excellent thread that is worth a look called "Ways to better air cooling".

In case you don't have any case fans at all, adding just one 140mm or 120mm intake fan in the front would make a huge difference, f.e.
 

·
Registered
Joined
·
103 Posts
Discussion Starter #17
yes, I'm going to have to educate myself about air-cooling. If I took out my disc drive up top and moved one of my hdd's further down I could put a third 120mm fan in front aligned with the cpu.

I'm a little worried now about the CPU itself. Is there a possibility that running it super hot with insufficient cooling for 4 months could have damaged the CPU itself?
I'm wondering whether I should RMA it. AMD shows a picture of a black spot on the underside of a CPU as something they will not accept. If I check my CPU and don't see any visible damage is it safe to assume the CPU is fine?
 

·
Master of Black Snow
Joined
·
19,338 Posts
Quote:
Originally Posted by RepCom1207 View Post

yes, I'm going to have to educate myself about air-cooling. If I took out my disc drive up top and moved one of my hdd's further down I could put a third 120mm fan in front aligned with the cpu.

I'm a little worried now about the CPU itself. Is there a possibility that running it super hot with insufficient cooling for 4 months could have damaged the CPU itself?
I'm wondering whether I should RMA it. AMD shows a picture of a black spot on the underside of a CPU as something they will not accept. If I check my CPU and don't see any visible damage is it safe to assume the CPU is fine?
It should still be within RMA period. When submitting, simply let them know what hardware you're running, the steps you took to correct the issue and the issue itself. So long as the CPU is POSTing I don't see them refusing the RMA. They've never done it to me and they've gotten quite a few from me from Athlons to FX chips.
thumb.gif


~Ceadder
smil3dbd4e4c2e742.gif
 
1 - 18 of 18 Posts
Top