FX 6300 Issues

[Cgermino89]

Hi there. I'm at my wits end with my sisters PC. Last year I built her a PC with a FX 6300 CPU MSI 970 Gaming Motherboard, and 8gb of DDR3 1866 Ram. I have a hyper 212 evo cooler on it, and a evga 500w power supply, and a GTX 750TI SC. Everything has been fine up until a few weeks ago. The CPU keeps reading at 1.33GHZ, and 2 core 4 threads in task manager. I called AMD up and went thru the RMA with them. I got my new CPU the other day, and reinstalled windows just to start off fresh. Well the CPU is doing it again, but this time task manager reads 3 cores, 6 threads like it should. I don't know what it could be at this point. My BIOS is set to the default optimized settings, and I've done everything I can with this PC. I'm hoping that someone here has had this issue, and knows a fix for this.

[it_dont_work]

Have you got ultra power saving disabled in the bios? I'll have to reboot to find the exact name

 

EDIT: PowerNow is what its called.

[Cgermino89]

8 minutes ago, it_dont_work said:

Have you got ultra power saving disabled in the bios? I'll have to reboot to find the exact name

 

EDIT: PowerNow is what its called.

where would this option be found in the BIOS?

[it_dont_work]

Just now, Cgermino89 said:

where would this option be found in the BIOS?

Probably somewhere in hardware and clock settings, don't know I've never used an msi bios. It's a setting that puts the cpu in a low p-state to save power and underclocks and undervolts the cpu.

 

It should auto boost the cpu clock when you do cpu intensive tasks though, so easiest way to check if this is the problem is to download cpu-z run the inbuilt stress test and watch what the clock rate does in the main tab.

[Cgermino89]

okay. I'll try that out and get back with you. Thanks.

[You_are_a_cunt]

Disable AMD Cool'n'Quiet.

 

Your CPU seems to be clocking down to save power.

 

Even though it's marketed as a 6-core, it's not actually a 6-core. It's 3 2-core "modules" and each of the 2 cores per "module" shares some interconnect.

 

My 4300 shows up as a 2-core 4-thread CPU, which is how it should look like.

 

Quick edit: Get Live Update off of MSI's website (check downloads on motherboard page) and update your bios (and whatever motherboard drivers are out of date).

 

 

(I have the same motherboard and I've not had any problems in the year and a half I've been using it)

[Cgermino89]

13 minutes ago, it_dont_work said:

Probably somewhere in hardware and clock settings, don't know I've never used an msi bios. It's a setting that puts the cpu in a low p-state to save power and underclocks and undervolts the cpu.

 

It should auto boost the cpu clock when you do cpu intensive tasks though, so easiest way to check if this is the problem is to download cpu-z run the inbuilt stress test and watch what the clock rate does in the main tab.

tried the CPUZ test and all it got up to was 1.37ghz and 40% utilization 

[it_dont_work]

That is a little od but as @revsilverspine said it's trying to stay in a power saving state, turn off cool and quiet and hunt around in the bios for any settings that look to be power related, then google them on your phone.

[incarnate]

Check your Windows power plan in the control panel while you're at it.

 

If it isn't a power saving plan causing it, it might be the vrms overheating

[Cgermino89]

what do I do about vrms overheating?

[SpaceGhostC2C]

3 hours ago, Cgermino89 said:

I got my new CPU the other day, and reinstalled windows just to start off fresh. Well the CPU is doing it again

Doing what exactly?

2 hours ago, revsilverspine said:

Disable AMD Cool'n'Quiet.

 

Your CPU seems to be clocking down to save power.

 

In which case the solution isn't disabling Cool'n'quiet, but just stop worrying about it  I mean, a CPU saving power while idle isa  good thing, isn't it?

2 hours ago, Cgermino89 said:

tried the CPUZ test and all it got up to was 1.37ghz and 40% utilization 

But you don't have one clock, you have 6. 40% utilization isn't enough to fire up all cores, so some of them could perfectly stay at power-saving frequencies. Or is that the max among the 6?

Try to describe better the situation and the symptoms, it will help coming up with diagnostics and potential solutions  

[Cgermino89]

Just now, Strike105X said:

RMA the motherboard (if its still in warranty), that one should be able to take in an FX8350 with a mild overclock, so a stock FX6300 for sure shouldn't be an issue. 

The warrenty is up, so purchasing a new one is the way to go then. Do you have any suggestions on a good motherboard for the fx 6300?

[Cgermino89]

11 minutes ago, SpaceGhostC2C said:

Doing what exactly?

In which case the solution isn't disabling Cool'n'quiet, but just stop worrying about it  I mean, a CPU saving power while idle isa  good thing, isn't it?

But you don't have one clock, you have 6. 40% utilization isn't enough to fire up all cores, so some of them could perfectly stay at power-saving frequencies. Or is that the max among the 6?

Try to describe better the situation and the symptoms, it will help coming up with diagnostics and potential solutions  

I've stressed it with cpuz benchtest program for all cores, and it only maxes out at 40% at 1.37ghz. even when it does hit say 80% or so the frequency doesn't change.

[SpaceGhostC2C]

2 minutes ago, Cgermino89 said:

I've stressed it with cpuz benchtest program for all cores, and it only maxes out at 40% at 1.37ghz

Can you use HWMonitor or any other program that reports all 6 clock speeds and utilizations? I believe CPU-z shows the highest of the 6, but just to make sure we aren't drowning in a glass of water.

If confirmed that even cores currently under load stay at ~1.4GHz (the power-saving frequency for FXs), then it does seem like a motherboard VRM issue. We can then check if temperature is the problem if you have some way to measure the temps on their heatsink. If not, we'll have to figure out something else.

3 minutes ago, Strike105X said:

PS: damn i miss Space Ghost and his talk show :(.

I miss Zorak even more, but that nick was taken  

[Cgermino89]

Okay

10 minutes ago, Strike105X said:

Okay my advice at this point would be:

 

1 Try doing a prime95 test and see if it still does it, use Hwmonitor to check things out like Space Ghost said. Also regarding VRM temps, usually if VRM's are to hot to touch, or the heatsink in this case is to hot, or the back of the wall on which the mobo is fixated is hot right where the VRM's are then yes they are definitely a problem, but it could also be that they have trouble outputting current and throttle, due to age or a malfunction. It could also be a PSU cable being the issue, which is why he should check out the PSU first. 

2 tell us your PSU, depending on what it is it might be worthwhile to borrow one and see if it still does it.

 

If the issue still persists it is definitely the Mobo, The best bang for the buck is the Asus 970FX gaming aura, if your tight on a budget i would recommend the M5A97FX R2.0 also by asus.

 

Damn the nostagia value, he was a great deal of fun indeed.

 

Edited with a few more details.

MY PSU is a EVGA 500W Non Modular power supply. I bought it brand new off amazon last year.

 

so i took a screenshot of the system idle and under cpuz stress test

 

[SpaceGhostC2C]

12 minutes ago, Cgermino89 said:

Okay

MY PSU is a EVGA 500W Non Modular power supply. I bought it brand new off amazon last year.

 

so i took a screenshot of the system idle and under cpuz stress test

So it was as bad as it seemed  

The system is stuck at power saving speeds, although there is no indication of thermal throttling. It could be VRM overheating (provided no mobo sensor is close enough to the VRMs) or overcurrent protections (if for example there is a leak or something shorted in the VRM area). Or just VRMs failing. Given the information on previous posts, most likely it'e the motherboard one way or another.

If it was just heat, though, it could be the VRM heatsink no longer making good contact. You can consider re-seating it, perhaps replacing the thermal pad (never did that myself), or making sure it didn't get loose somehow (don't over-tighten it though). You can also place a small fan on top of the heatsing to help dissipate the heat. But if it's not a mechanical issue with the heatsink, the question would remain: why did it start getting hot only now? The answer could still be worrisome, even if you manage to get your CPU to work at higher speeds.

Testing on a different motherboard would be ideal.

[Cgermino89]

15 minutes ago, Strike105X said:

Like i said try Prime95 for stress testing just to be sure, though at this point yes, if its nothing related to the PSU then the mobo is the culprit. Those two mobo's i listed are reliable, so you shouldn't worry, though to be honest you just had bad luck, the MSI 970 Gaming motherboard is one of the few actually good motherboard from MSI.

I'll do a prime95 test when I get home. Any suggestions to what settings I should use on prime 95

2 minutes ago, SpaceGhostC2C said:

So it was as bad as it seemed  

The system is stuck at power saving speeds, although there is no indication of thermal throttling. It could be VRM overheating (provided no mobo sensor is close enough to the VRMs) or overcurrent protections (if for example there is a leak or something shorted in the VRM area). Or just VRMs failing. Given the information on previous posts, most likely it'e the motherboard one way or another.

If it was just heat, though, it could be the VRM heatsink no longer making good contact. You can consider re-seating it, perhaps replacing the thermal pad (never did that myself), or making sure it didn't get loose somehow (don't over-tighten it though). You can also place a small fan on top of the heatsing to help dissipate the heat. But if it's not a mechanical issue with the heatsink, the question would remain: why did it start getting hot only now? The answer could still be worrisome, even if you manage to get your CPU to work at higher speeds.

Testing on a different motherboard would be ideal.

I even put a box fan on top of the open case to see if the speed of the CPU would change and nothing happened. I'll definitely do one more stress test and if nothing happens. I'll buy a new mobo or see if I can borrow one.