Low temps while stress test?

[Brian]

Hi guys, just got my board in yesterday and the system is up and running now!
I have a couple of questions quick though. Yesterday when I opened up HWMonitor it read all 4 cores on my Q9550, but one core was at times 12 celcius cooler than the rest. Is that a problem? I know others have posted, but the higher temp cores aren't in the really high temperatures so I was wondering if anything's wrong with that one core?
Here's a pic:


Also, when I started up HWMonitor just a few minutes ago, it only read 3 cores. Once I started up Prime95 (blend), I restarted HWMonitor to see if that core #0 was just dormant for the time being because it wasn't being used. When I restarted the program, it read all 4 cores again.

Pic with HWMonitor reading 3 cores:


Pic with 4 cores:


Only 1 core really seems to be under load judging by the temperatures...that doesn't seem normal to me?

One last thing, HWMonitor also only reads 1 core of my GPU as well. I know I'm putting a lot of faith into a piece of software, but does the new version only read the higher temp core? I'm hoping there's nothing wrong with the system and it's just a software thing.

Thanks.

[b1lk1]

Try CoreTemp or RealTemp and see if the same things happens.

[Brian]

OK, CoreTemp and RealTemp both read 4 cores and both read the same temperatures.
So with that said, the only thing that concerns me is the HWMonitor only seeing 3 cores? Core #0 is the coolest one and that's the one HWMonitor doesn't see.
I did a quick run of Prime95 and it passed twice on all 4 cores, so I don't know why it's not reading the 4th core.

Edit: it seems HWMonitor will recognize all 4 cores when the system is under a bit of a load when the program is launched. I looked in the task manager for Vista under Performance and when I hit start on Prime95 all 4 CPU Usage History line graphs go up to the top and it says it's running at 100%. Just want to run it by someone on here and see if I should be concerned that maybe one of my cores isn't working? My concern comes from that Core #0 that isn't being read on HWMonitor being at 16 celcius while the other 3 cores are at 30 celcius. I've seen differences in temperatures between cores before, but never 14 celcius different. That's why I'm wondering if the core is idle or dormant while not being used and thus cooler or if it just isn't working properly.

[SugarJ]

I find HWmonitor is handy for drive temps and the like, but I stick to CoreTemp or Realtemp for core temps, they are more accurate. And I use GPU-Z for gpu temp monitoring.

[enaberif]

and are you running SmallFFT?

[Brian]

Good advice, I'll be using those mentioned programs from now on, but I just want to make sure that the core #0 is working properly. Note my last post edits (sorry came in late) about Prime95 and the CPU usage through task manager. The core does get hotter when playing a game or doing something under load, just again concerned as to if it is working properly due to the big difference in temperatures.

[SugarJ]

As long as you are seeing the temps change with load, and the one core is lower than the other 3, I wouldn't be too concerned. It's when you have one core way higher than the others that you have a problem as it limits your overclock.

And try OCCT 3.0 for your stability testing too. Don't be surprised with the Linpack temps, they are usually 10 degrees higher than anything you'll see under normal use.

[Snafu]

It could be that the sensor on teh one core is off a bit from the others. You can always recalibrate RealTemp.

[3.0charlie]

Or that your chip has 3 stuck sensors. Use OCCT or Linpack to really push the chip in the upper 60s.