Question Disappointing FX 9590 temps w/ Noctua NH-D15

[discourse4]

If any of you wants the genuine fx9590 experience, do not use the NH-D15. I somehow cant get it over 45c even with Prime95 running for 26 hours. Granted, it was outside of a case.

I was looking to challenge myself to air-cool this, so the temp is beyond underwhelming. I was planning a 9x 120mm fan case set-up, but I will have to put this project on hold until I can get a worse cooler.

It didn't even crash once.

 

[discourse4]

That means that 70% of the worlds 9590s are also thermal throttling. This is actually both extremely funny and also realistic - meaning that I want to prove this to be true so badly.

Anyone know of any other way to prove I am being throttled? Aida64 is not showing drops in usage or clocks after ~20 minutes.

 

[Gam3r01]

Every single core reads the exact same thermal margin, current, average, peak, and minimum.
Thats not just unlikely, thats impossible.

 

[discourse4]

Every single core reads the exact same thermal margin, current, average, peak, and minimum.
Thats not just unlikely, thats impossible.

The average looks different to me. Unless you are expecting variance of at least 1C.

 

[Gam3r01]

A difference of .01C is a monitoring error, not an actual difference.
You should expect a several C difference between cores on average, just given the layout of the processor itself.

Thats to say nothing of the current, min, and max values being identical as well.

 

[discourse4]

A difference of .01C is a monitoring error, not an actual difference.
You should expect a several C difference between cores on average, just given the layout of the processor itself.

Thats to say nothing of the current, min, and max values being identical as well.

Can you explain this more thoroughly? It seems counterintuitive.

This is a 20 minute average of the thermal margin, polled every 1s. For this to have a 1C difference, it would mean that I would have to see my cores at different temps at ALL times (this would mean 1 core stuck at 25, another at 24, etc.)

I also dont understand the issue with Min & Max. I always thought these are the temporary spikes and outliers. I would need more info for this, can you post an expected min/max from the AMD overdrive?

 

[Gam3r01]

You have 8 cores that all have the exact same outlier peaks, thats not possible.
This is what you should expect a run to look like:
Just ran on my system: https://gyazo.com/d1e173b226949c01e892095ef034effc

 

[discourse4]

HWMonitor is not reliable for FX processors. It also doesn't support a per core reading of the temp for the processor. I posted a pic yesterday of my HWMonitor temps and was told (and I agree after research), that only AMD Overdrive should be used for FX temps.

On that point I am having trouble finding results online that shows the cores being wildly non-uniform on AMD Overdrive. Too few images to make a conclusion.

https://www.google.com/search?rlz=1...0..0.0.0.......0......gws-wiz-img.CQrGWPxOJi8

 

[Gam3r01]

Read my sig, I said it was on my system, which is not an FX system.

 

[discourse4]

Have you used AMD Overdrive before then? Because I accomplished your claim of impossibility twice.

This is on an a phenom 840t after 5 minutes.

EDIT: this is on a different MSI motherboard. Also, this was only after 3 minutes. I sent the wrong pic to my email.