• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Can Someone Please Check My Temps? Phenom II & Thor's Hammer

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

sweefu

Member
Joined
Oct 15, 2009
Location
Canberra, Australia
Hey all,
today i picked up my Xigmatek Thor's Hammer cpu cooler, i installed it with standard cooler master thermal compound. I did a 10minute run with prime 95 blend test and these are the temps i got:

Idle Before Test: 25-27 degrees C

Full Load 10min Test: 38degrees C

Ambient: 20-25 degrees C (more or less a guess)

The CPU is a Phenom II 550BE and is stock.
I am using a thermaltake M9 case with 2x120mm intake, 1x120mm exhaust side, 120mm exhaust rear, 120mm exhaust through PSU and 1x120mm fan on top of the cooler exhaust towards PSU fan.

How do these temps sound?

Thankyou.
 
Last edited:
Thats what it seems to be haha. thermometer up higher than my computer though, computer is on the floor.
I'm using CPUID hardware monitor, here is a screen shot.

So temps are good?
Cheers
 

Attachments

  • Untitled22.jpg
    Untitled22.jpg
    94.1 KB · Views: 186
if theres fans on the thors it will cool below ambient.. why not.. fans cool the air somewhat on pushing/pulling..

temps look great.. whats the worry?
 
Just wanted to check, i was a bit worried about the full load temp but if it isnt bad then that is good. Tomorrow i will be putting another fan on the hs to see if the push/pull goes any better.

Is the load temp good or sort of acceptable?

Thanks.
 
Unless you are using some kind of active cooling measure such as phase change or TEC or a bong evaporative loop, there is no way to get cooler than ambient temps on the processor.

But those temps are OK, so you should be good to go with doing some overclocking.
 
Some O'Cing done

Hey, thought i'd post up my first results

vCore: Stock
Multiplier: 16.5
H.T Ref. Clock: 210
Core Speed: 3.465Ghz

Ambient Temp: 25 degrees C

Idle Temp: 28 degrees C

Full Load Temp: 40 degrees C

What does everyone think?
Any tips? - This is more or less my first time properly overclocking.
Cheers,
Anthony.
 

Attachments

  • Untitled3.jpg
    Untitled3.jpg
    109.2 KB · Views: 164
First - great temps :).

if theres fans on the thors it will cool below ambient.. why not.. fans cool the air somewhat on pushing/pulling..
Thermo-nerd here... figured I might try clarifying so you can ace your next thermodynamics course ;).

CPU's/heatsinks will never be cooled below ambient drybulb* temperatures if they're only being cooled by ambient air, moving or not. Never Never Ever**.

Fan motors dump heat into the air stream, which actually warms the air stream. It's a significant effect in larger fans such as those used designing building HVAC systems, but something we generally disregard on the scale of a computer fan and it's effects on a CPU.

By virtue of being in forced motion, air will heat/cool a surface (such as heatsink fins) at a different rate than otherwise still air, but ambient air will heat up any surface that falls any lower than ambient temps.

In summary:
  • Surface temp = Ambient temp: no heat transfer occurs.
  • Surface temp < Ambient temp: air will heat up the surface.
  • Surface temp > Ambient temp: surface will heat the air.

---

*Drybulb is what most thermometers/thermal sensors read. It's a measure of the heat energy in the air in isolation of any moisture. Since typical computer heatsinks and other cooling processes involve no moisture exchange (unless you want to get creative), assume any temperatures in the context of ocforums discussions to be in drybulb. What's important to the layman is that this is not the "feels like" or "wind chill factor" temperature you hear on the news.

**I mean it!
 
First - great temps :).


Thermo-nerd here... figured I might try clarifying so you can ace your next thermodynamics course ;).

CPU's/heatsinks will never be cooled below ambient drybulb* temperatures if they're only being cooled by ambient air, moving or not. Never Never Ever**.

Fan motors dump heat into the air stream, which actually warms the air stream. It's a significant effect in larger fans such as those used designing building HVAC systems, but something we generally disregard on the scale of a computer fan and it's effects on a CPU.

By virtue of being in forced motion, air will heat/cool a surface (such as heatsink fins) at a different rate than otherwise still air, but ambient air will heat up any surface that falls any lower than ambient temps.

In summary:
  • Surface temp = Ambient temp: no heat transfer occurs.
  • Surface temp < Ambient temp: air will heat up the surface.
  • Surface temp > Ambient temp: surface will heat the air.

---

*Drybulb is what most thermometers/thermal sensors read. It's a measure of the heat energy in the air in isolation of any moisture. Since typical computer heatsinks and other cooling processes involve no moisture exchange (unless you want to get creative), assume any temperatures in the context of ocforums discussions to be in drybulb. What's important to the layman is that this is not the "feels like" or "wind chill factor" temperature you hear on the news.

**I mean it!

Yupyup, all very true.

You can sneak around it with compressed air if you take the heat of compression out of it and then decompress it, but a fan doesn't do that (and we're talking a good bit of pressure here, too). That's more like a phase change setup though, except less efficient :p
The reason moving air feels colder to you is twofold, evaporation from your skin, still air has a thick boundary layer of moist warmer air (you skin has warmed and humidified it), moving air pushes that away, exposing your skin to ambient temp air (feels cooler then the skin-warmed air) and dry air, which sucks moisture out of your skin making your skin colder (phase change).
Computer heatsinks are not wet, and hence don't get the evaporative effect.
They still get the boundary layer effect, which is why we have fans on them in the first place.


AMD temp sensors aren't any better then Intel sensors at the low end, yours read lower then reality, leading to an almost-ambient cpu die being read as sub-ambient.

Regardless, those temps are excellent and you can (and should) continue with your overclocking, i don't think you'll have any heat issues at all with that CPU and HSF.
 
Why would you guys think his temps are fine if some of them are being reported as below ambient? Obviously something is not correct, therefore the other temperatures may be very off as well...
 
Why would you guys think his temps are fine if some of them are being reported as below ambient? Obviously something is not correct, therefore the other temperatures may be very off as well...

I seem to recall a report within the last year pointing out that the most a CPU's internal (core) sensor could realistically be off was within a range of 10*C. That said I just spent a few minutes in google and can't seem to find that article... anyone know what I'm talking about?

Personally I don't condone "pushing the limits" of any processor without using a means of temperature measurement you can verify outside the case. That said I assume a conservative 10 degree deadband and do utilize hardware temp readouts.

If his processor, reading 40 max is really at 50, I'd be comfy with that - would you?

You can sneak around it with compressed air if you take the heat of compression out of it and then decompress it, but a fan doesn't do that (and we're talking a good bit of pressure here, too). That's more like a phase change setup though, except less efficient :p
I very carefully chose my words by saying "only being cooled by ambient air." :)
 
I seem to recall a report within the last year pointing out that the most a CPU's internal (core) sensor could realistically be off was within a range of 10*C. That said I just spent a few minutes in google and can't seem to find that article... anyone know what I'm talking about?

Personally I don't condone "pushing the limits" of any processor without using a means of temperature measurement you can verify outside the case. That said I assume a conservative 10 degree deadband and do utilize hardware temp readouts.

If his processor, reading 40 max is really at 50, I'd be comfy with that - would you?


I very carefully chose my words by saying "only being cooled by ambient air." :)

I'm pretty sure that information of the 10* difference is older than 1 year, I've heard that about 4 years ago. Either way, from what I understood/recall, the only accurate temperature for the CPU should be taken from the on-die sensor temperatures, because sometimes the temps can be reported from off-die sensors, which are the ones that are way off. There's so many ideas out there, it gets confusing.

My first time accessing the on-die sensor on my system came from using CoreTemp.
 
hey guys just to clear something up quickly,
my first temps that i gave, i measured ambient temp with a thermometer on my wall, and my computer case was on the ground. obviously to start off with a mercury thermometer isnt very accurate, there is a room for error. Secondly, because the thermometer is up higher, it was a hot day heat rises and the computer case was on the ground so this secondly makes up for the 2 degree overlap in temps.

Also,
thanks very much for the thermal dynamics info :D i just finished thermo at school but we didnt cover much over this topic, mainly did heat pumps etc.

Thanks again guys.
 
Back