Different conflicting CPU temps.

[Dravonic]

I'm overclocking my FX 8350 but I found one unexpected problem, different monitoring programs will tell me different CPU temperatures, and I don't know what I should be taking into consideration. HWiNFO64 seems to be the most verbose, so I'll let you guys know what it tells me.

Under CPU[#0]: AMD FX-8350 I have a CPU 0 temp, and it's currently maxing out at 60C. But in another section ITE IT8721F I have a CPU temp, that goes as high as 68C.

Other monitoring programs don't seem to agree which temps to show in the CPU temp, some will show me the higher temp, others the lower. With some googling it seems I should be aiming for something between 60-65C, but now I'm confused. Which temp should I take into consideration? Do I have some more headroom for overclocking or do I need to tone it down a bit?

 

[Idontcare]

Have you tried Core Temp?

The other issue you will find is that there are no specs on the thermals for your FX-8350. Mine can go all the way to 83C (per CoreTemp) without throttling of any kind, but at 83C the system just shuts off (black screen), also without any throttling involved.

Seems like a mystery at the moment, but the concensus says don't OC your 8350 such that it routinely runs >60C when doing anything that isn't specifically a stress test like Prime95. I target 65C max while stress testing, which usually means I'm looking at nothing higher than 45-50C when running pretty much any other app.

 

[Jovec]

This conflict typically arises from programs reading the socket temp (a sensor on the motherboard under the CPU socket) instead of the core temp as reported by the CPU.

Unfortunately, AMD hasn't released thermal info for anything other than Phenom 2/Thuban. What we do know came from this AMD post and many seem to use it as guidelines for AMD's current generation of CPUs. This info may or may not apply to FX, Llano, and Trinity (it probably doesn't). For example, AMD recommended staying 28C below the thermal shutoff for Thuban; this same delta puts max FX temps at 55C for daily usage. Given the different process nodes and voltages between Thuban and FX, I have a hard time believing such inferences are accurate.

http://www.overclock.net/t/1128821/amd-temp-information-and-guide

Sorry I didn't get back to you sooner, I just recently had your email forwarded to my department.

Concerning your question regarding the temperatures with your processor. (1090) the maximum temperature threshold is 62 Celsius which set for the internal die (core) temperature of the chip. The core temperatures have an equational offset to determine temperature which equalizes at about 45 Celsius thus giving you more accurate readings at peak temperatures. The hindrance in this is the sub ambient idle temperature readings you speak of.

The silicon and adhesives used in manufacturing these processors has a peak temperature rating of 97+ Celsius before any form of degradation will take place. The processor also has a thermal shut off safe guard in place that shuts the processor down at 90 Celsius.

The Cpu temperature is read form a sensor embedded within the socket of your motherboard causing about a 7-10 Celsius variance form the actual Cpu temperature, which may be what you are reading about on the net.

You can use an application called AMD overdrive, that will allow you to monitor your temperatures accurately.

As long as your core temperature has not exceeded the high side of the 60 degree mark for extended periods of time you should be ok. 62 degrees holds a generous safety net to begin with.

I hope I was able to answer your questions, If you have any more inquiries don't hesitate to contact us.


Thank You

Alex Cromwell
Senior Technology Director
Advanced Micro Devices

If we assume AMD used some of the same implementations in FX, then:
1) Idle temps are still inaccurate
2) There is still no per-core temperature sensor
3) AMD recommends a large safety net between the thermal shutoff point and the recommended everyday usage temperature.
4) There also still doesn't appear to be any thermal throttling in FX; instead, it merely shuts down (observed by IDC's post)

AMD has recommended using either AMD Overdrive or CoreTemp. Presumably any other monitoring program with results that match either of these could also be accepted as working.

 

[Dravonic]

Core Temp will give me the same temp in HWiNFO64 under CPU[#0]: AMD FX-8350.

Either way, should I just ignore the socket temp? It won't stop going up, meaning I nearly have an overheating problem even at stock clocks and voltage.

My cooler is a Corsair H70, and I'm hitting 68C(core) @4.6 and 1.42v in prime95 in less than 5 mins. I pulled the plug. Other tests like AIDA64 won't go past 65C. Should I look at replacing the stock thermal paste? Or lapping the pump's contact plate? Or even get a new cooler?

 

[Idontcare]

Either way, should I just ignore the socket temp? It won't stop going up, meaning I nearly have an overheating problem even at stock clocks and voltage.

I can't say with any authority on the matter that it is OK to ignore socket temp, but I can say that I have always ignored the socket temp and nothing bad has happened to my hardware from doing so.

But you mention the value creeps upwards even at stock clocks and voltage, which makes me wonder if you have some other hardware issue going on? Do you have decent air cooling going on at the mobo surface?

If you just angle a case fan at the socket area, hold it by hand for a quick test, while monitoring the socket temp does the socket temp decrease?

 

[Dravonic]

Due to the shitty nature of my case, I have to leave it open or it gets much worse. Since it's open the lack of proper airflow through the case might be the problem, but I don't think it should be much worse. Or is it?

 

[guskline]

I use AIDA64 sensor for my temps.