5600x running at 96C

[Abhigunner]

Hi, I have a 5600x paired with a Noctua U12S since 2020 and whenever I am gaming it touches 96C within 5 mins and stays there. I have tried repasting and cleaning the fans but to no effect.

My understanding is that 5600x being a 65w should be cooled adequately by the noctua cooler. Does anyone with a 5600x or this combo has this issue.

Also my motherboard is a x570 Aorus Master and I was wondering whether it is worthwhile to upgrade to a 5700x3d as I dont want waste a good motherboard by upgrading to a 7000 or 9000 series. My GPU is a RTX 3080FE
Also planning to repaste my 3080 fe so if anyone has any advice or good guides please help

 

[bssunilreddy]

Hi, I have a 5600x paired with a Noctua U12S since 2020 and whenever I am gaming it touches 96C within 5 mins and stays there. I have tried repasting and cleaning the fans but to no effect.

My understanding is that 5600x being a 65w should be cooled adequately by the noctua cooler. Does anyone with a 5600x or this combo has this issue.

Also my motherboard is a x570 Aorus Master and I was wondering whether it is worthwhile to upgrade to a 5700x3d as I dont want waste a good motherboard by upgrading to a 7000 or 9000 series. My GPU is a RTX 3080FE
Also planning to repaste my 3080 fe so if anyone has any advice or good guides please help

The following might be happening:

Not enough contact by CPU IHS with cooler block,
Mobos VRMs not working properly or heating up
CPU itself is problematic
Check if the cooler fans are working or spinning
Not enough positive air is circulating in the cabinet

Did you remove the plastic cover that comes with coolers undersurface

My 5600x never crosses 75° celsius even while intense gaming. I am using Deepcool AK400 with push pull setup. Fans are Arctic P12 PST PWMs.

 

[Tracer_Bullet]

yeah is the gpu newly installed ?
Do you get high cpu temps without gpu load, ex using prime95 or OCCT cpu tests.

If not, then probably your cabinet is unable to deal with the heat removed from the gpu. I had this problem with an old style CM cabinet + 3080 +3300x. Mesh style cases can have very good airflow to deal with such issues.

 

[Abhigunner]

Mobos VRM temps are fine.
Cpu May be faulty but even at 96c it boosts to 4.5 Ghz.
I have ordered an AIO to check whether there is some issue with the noctua cooler.
Air flow may be an issue but should it cause the temps to rise this much?

yeah is the gpu newly installed ?
Do you get high cpu temps without gpu load, ex using prime95 or OCCT cpu tests.

If not, then probably your cabinet is unable to deal with the heat removed from the gpu. I had this problem with an old style CM cabinet + 3080 +3300x. Mesh style cases can have very good airflow to deal with such issues.

GPU has been installed since the start. Case is an NZXT 440. Do you think I need to change it?
Will try Prime95 and get back to you

 

[Heisen]

Are you overclocked? Or did any change related to that? Try resetting bios.

96c in itself doesn't tell the complete story, the correct way to compare is get your ambient temperature and subtract it from your CPU max temperature, then you get the delta, total rise in temperature.

This is the temperature you will compare with others to get accurate measurements.

Also comparing idle temperature means nothing, because it is basically your ambient temperature.

To check if cabinet is an issue, just open both the side panels, if you see major change in your max temperature then case is the issue otherwise you can rule it out.

Also planning to repaste my 3080 fe so if anyone has any advice or good guides please help

Why open a perfectly torqued down GPU, when it is not giving you any problems, or thermal throttling aggressively.

 

[Abhigunner]

Are you overclocked? Or did any change related to that? Try resetting bios.

96c in itself doesn't tell the complete story, the correct way to compare is get your ambient temperature and subtract it from your CPU max temperature, then you get the delta, total rise in temperature.

This is the temperature you will compare with others to get accurate measurements.

Also comparing idle temperature means nothing, because it is basically your ambient temperature.

To check if cabinet is an issue, just open both the side panels, if you see major change in your max temperature then case is the issue otherwise you can rule it out.


Why open a perfectly torqued down GPU, when it is not giving you any problems, or thermal throttling aggressively.

The GPU has started throttling massively, goes to 88C and Clocks down to base 1440 Mhz in 5 mins.

My thinking was that both of them being at a high temperature might be due to one or the other so tackle both.

 

[Abhigunner]

Also no Overclocking, just PBO enabled

 

[Tracer_Bullet]

GPU has been installed since the start. Case is an NZXT 440. Do you think I need to change it?

No, if it was fine before with the gpu, then that is not the cause.
Maybe check that all fans are working.

Still test the cpu without gpu load and that will confirm that cpu has a problem getting cooled.
But it does seem that way, have never faced this.

Until you figure out the reason, You can try to set a temp/power limit in bios ( maybe ryzen master too) and check benchmarks to see what perf loss you see. Last bits of performance usually takes too much power/creates heat so we may not have to give up too much perf.
You can try eco mode to keep it simple.

Also planning to repaste my 3080 fe so if anyone has any advice or good guides please help

Unless you have an issue and know what you are doing, i wouldn't take the risk.

 

[chetansha]

Did you remove the plastic thingie from the cpu cooler where it contact the cpu

 

[Abhigunner]

Did you remove the plastic thingie from the cpu cooler where it contact the cpu

Yeah, first thing I checked, and have repasted twice.

 

[ch@ts]

Your CPU is designed to keep boosting it's clock speeds till it hits it's max temperature, ie ~ 95 degrees. So, it is running within spec.
I've got a 5500GT and that used to run at max clocks, and max speeds till it touched 95, but that was with the stock cooler, but then it would throttle a bit. I don't think your CPU is throttling. I installed a Cooler Master Hyper Tx3 and the max temp dropped to 82, and the throttling ended.

You mentioned that you've had the CPU since 2020. Have the temps always been this high? What are your idle temps? What's the airflow in the cabinet like? How hot does the rest of your system get? Why do you want to repaste your GPU? How hot does it run? Also, as @Heisen said, your system temps depend on the ambient air temperature. What is that like?

 

[Abhigunner]

I habe

No, if it was fine before with the gpu, then that is not the cause.
Maybe check that all fans are working.

Still test the cpu without gpu load and that will confirm that cpu has a problem getting cooled.
But it does seem that way, have never faced this.

Until you figure out the reason, You can try to set a temp/power limit in bios ( maybe ryzen master too) and check benchmarks to see what perf loss you see. Last bits of performance usually takes too much power/creates heat so we may not have to give up too much perf.
You can try eco mode to keep it simple.


Unless you have an issue and know what you are doing, i wouldn't take the risk.

I have checked with occt and idle temps are hovering around 65c, with 89c on load in occt cpu test.

Your CPU is designed to keep boosting it's clock speeds till it hits it's max temperature, ie ~ 95 degrees. So, it is running within spec.
I've got a 5500GT and that used to run at max clocks, and max speeds till it touched 95, but that was with the stock cooler, but then it would throttle a bit. I don't think your CPU is throttling. I installed a Cooler Master Hyper Tx3 and the max temp dropped to 82, and the throttling ended.

You mentioned that you've had the CPU since 2020. Have the temps always been this high? What are your idle temps? What's the airflow in the cabinet like? How hot does the rest of your system get? Why do you want to repaste your GPU? How hot does it run? Also, as @Heisen said, your system temps depend on the ambient air temperature. What is that like?

Temps have been on the higher side only. Whenever I had monitored it was around 90c only.
Only bothering now because now gpu is also throttling I believe because the gpu Temps also rise to 87c and then the clock goes from 1780 Mhz where it starts to 1440 Mhz after 10 minutes and the tdp also drops to 75 percent. I have observed this recently while playing star wars outlaws. Read online that 3080 fe had bad thermal pads and hence thought to repaste.
I have ordered an Arctics Freezer iii 360 and will check with that.

 

[roadrash99]

AM4/AM5 will always try to hit 95 degC. If you get a better cooler like a 360mm AIO it will still hit 95 degC just at a slightly higher clock. As long as you are hitting the advertised clocks this temp is not a concern.

If you really want a lower temp set a power limit or temp limit in bios

 

[Abhigunner]

Temps after occt for 13 mins

 

[Heisen]

just PBO enabled

This, why do you have this enabled? It is overclocking because it causes the CPU to run out of it's stock settings. There is a reason why it's disabled by default.
Your 96c is fine behavior, because you have basically asked it to overclock itself as long as it has the thermal budget.

• PBO is an optional automatic overclocking feature that allows AMD Ryzen processors to boost higher and longer by relaxing power and thermal limits.
• By default, AMD CPUs operate within standard power and thermal specs (e.g., TDP, PPT, TDC, EDC) to ensure stability and longevity.
• Enabling PBO allows the CPU to exceed those limits as long as thermals and power delivery allow it, giving you better performance in workloads that benefit from higher boost clocks.

 

[Abhigunner]

Ok, will disable pbo and test. Thanks for taking the time to answer

This, why do you have this enabled? It is overclocking because it causes the CPU to run out of it's stock settings. There is a reason why it's disabled by default.
Your 96c is fine behavior, because you have basically asked it to overclock itself as long as it has the thermal budget.

 

[YeAhx]

If you ask me, considering your GPU is also thermal throttling could it be your case is not getting fresh air and is choked ?
Also why not run an slight undervolt on both processor and the GPU? If you do not wanna dillydally just set a negative all core value of -20mV for processor and put GPU at 90% power limit (it will not exactly be an "undervolt" but it should stop the GPU from running on OC setting set by its vbios, mainly seen on OC edition cards)

Your CPU is designed to keep boosting it's clock speeds till it hits it's max temperature, ie ~ 95 degrees. So, it is running within spec.

Isn't this only true for when PBO is enabled?

Temps after occt for 13 mins
View attachment 229547

78W of power draw for a 5600x is high. I assume it's because of PBO pushing it aggressively so things should be better now that you have disabled it. Can you run a cinebench all core benchmark? Do this while running HWiNFO64 in the background in summary mode and lemme know the results of both.

BTW 5700X3D is exactly what you want if you want to upgrade just the processor. I have it for my PC (5700X3D + 3070) and it runs at 55-60C while gaming. I have it running stable undervolted at all core -30.

 

[Abhigunner]

If you ask me, considering your GPU is also thermal throttling could it be your case is not getting fresh air and is choked ?
Also why not run an slight undervolt on both processor and the GPU? If you do not wanna dillydally just set a negative all core value of -20mV for processor and put GPU at 90% power limit (it will not exactly be an "undervolt" but it should stop the GPU from running on OC setting set by its vbios, mainly seen on OC edition cards)


Isn't this only true for when PBO is enabled?


78W of power draw for a 5600x is high. I assume it's because of PBO pushing it aggressively so things should be better now that you have disabled it. Can you run a cinebench all core benchmark? Do this while running HWiNFO64 in the background in summary mode and lemme know the results of both.

Will do and get back to you.
PBO was set to auto. I have disabled it but still the same results.

PBO was set to auto. I have disabled it but still the same results.

 

[YeAhx]

It has to be the high power usage what's causing heating problem. Something somewhere must still be applying an OC on the processor. You did not install and use Ryzen master right? Your best bet would be to go to bios and turn on Eco mode for processor (google it how to do it) just to try and see the results. Since 5600x is a 65W TDP and Eco mode will run it at 45W max you will see better results. You might even gain performance doing this since it will not be thermal throttling and clocks will boost higher consistently. Kill two birds with one stone kinda thing.

Edit: forgot to say this, apply an undervolt if you haven't yet. Start with -15 on all cores then go to -20 or even -25 if no errors are reported (OCCT test is perfect for this one but prime95 is an option as well).

 

[Abhigunner]

It has to be the high power usage what's causing heating problem. Something somewhere must still be applying an OC on the processor. You did not install and use Ryzen master right? Your best bet would be to go to bios and turn on Eco mode for processor (google it how to do it) just to try and see the results. Since 5600x is a 65W TDP and Eco mode will run it at 45W max you will see better results. You might even gain performance doing this since it will not be thermal throttling and clocks will boost higher consistently. Kill two birds with one stone kinda thing.

Edit: forgot to say this, apply an undervolt if you haven't yet. Start with -15 on all cores then go to -20 or even -25 if no errors are reported (OCCT test is perfect for this one but prime95 is an option as well).

Thanks for the detailed reply. Will do all the steps and get back to you tomorrow
Also I should use ryzen master for applying offsets, right?