Tom's Hardware > Forum > CPU & Components > CPUs > Core 1 10.C over core 0?

Core 1 10.C over core 0?

Forum CPU & Components : CPUs - Core 1 10.C over core 0?

Tom's Hardware: Over 1.4 million members in 6 different countries available to answer all your high-tech questions. Sign up now! Its free!
Word :    Username :           
 
vibe   11-06-2007 at 01:20:50 PM

Hi all,

I have an AMD 4800+ X2 and was wondering if this is normal.

Under load test with Orthos my core 1 is in the high 50s and core 0 in high 40s at idle core 0 is 30/31 and core 1 35/36

I've had a good look at my HS and far as I can tell its fitted ok, is this normal or should I be looking for a fault.

V

Sponsored Links
Register or log in to remove.
epsilon84   11-06-2007 at 01:45:27 PM
- 0 +

There shouldn't be that much difference between the cores, generally they should be within 1C of each other under identical loads.

What software are you using to read the temps?

Reply to epsilon84
vibe   11-06-2007 at 01:55:23 PM
- 0 +

Core temp 0.95 and speedfan 4.33

they are giving different readings within 2 or 3 degs of each other but the difference is always similar.

Reply to vibe
epsilon84   11-06-2007 at 02:15:25 PM
- 0 +

Try CoreTemp 0.95.4 beta or Speedfan 4.34 beta, both have more accurate readings than previous versions.

Reply to epsilon84
rodney_ws   11-06-2007 at 02:21:13 PM
- 0 +

Perhaps unevenly applied thermal paste? Man, that's still a big difference... let us know what you find out. Never read a post like this before (not that I've read EVERY post)

Reply to rodney_ws
akhilles   11-06-2007 at 03:10:38 PM
- 0 +

Is your AMD 4800+ X2 by any chance a Brisbane?

Reply to akhilles
vibe   11-06-2007 at 08:49:26 PM
- 0 +

No it's a Toledo, why?

I'm wandering if the temp reader on CPU may be damaged, core 1 drops in temp very rapid after load, like in a 10th of a second, where as core 0 cools over a few seconds, any way to test? can anyone recomend an after market temp guage for CPU?

Reply to vibe
coret   11-06-2007 at 09:43:28 PM
- 0 +

Have you checked with a razor blade to see if the mating surface of the heatsink is actually flat? If it's not, then the CPU won't be evenly cooled even if the thermal gunk appears to be squished properly.

I bought the low profile arctic cooling HSF recently, and when it arrived it quite badly needed lapping to make it flat (i knew there was a good chance it would need it thanks to frostytech) but once it was done, the temps settled down on my 6000+ and it was then actually able to run orthos for more than 3 minutes (12 hour stable ... good enough for me)

Reply to coret
damaster   12-24-2007 at 06:17:52 AM
- 0 +

I'm using a Scythe Mini Ninja with Arctic Cooling MX-2 thermal paste on an AMD Athlon 64 X2 6400+ (Windsor). CoreTemp and SIW are both showing a 5-6 degree difference between my core 0 and core 1 temp. At idle core 0 is 37 and core 1 is 31.

Is this due to unevenly applied thermal paste? Or is the heatsink not installed flat?

Thanks

Reply to damaster
skittle   12-24-2007 at 06:49:30 AM
- 0 +

I have a lapped (800grit) scythe ninja and a lapped 3800+ @ 2.4ghz. core #1 runs ~1-3C higher than core #0.

10C is quite a lot though

Reply to skittle
nowwhatnapster   05-31-2008 at 08:53:46 AM
- 0 +

sounds like your heatsink is on crooked. would make the most sense. double check your setup.

To test: clean HSF & CPU completely of all thermal compound. apply a tiny amount of compound to the HSF only. Spread it evenly. wafer thin amount, just enough to stick to another surface to visibly see that they have made contact. Re-mount your HSF to your CPU, then take it back off. Check to see if the thermal paste that touched the CPU is evenly distributed. If it is. then.. thats not your problem and it sounds like its a problem with the CPU sensors IMO. Rma it. or if your lucky enough to have another dual core CPU on hand, test it in your system.

Reply to nowwhatnapster
skypod1   02-22-2009 at 03:50:15 AM
- 0 +

Old thread I know, but fwiw I have a similar situation with dual E5472 quad-core Xeons in an Asus DSEB-DG motherboard. Core temps according to SpeedFan 4.37 show a similar 10C variation pattern on both CPUs:
Core 0 50C
Core 1 60C
Core 2 43C
Core 3 43C

Core 4 38C
Core 5 48C
Core 6 44C
Core 7 48C

Using SIW, core temperatures also show the same pattern but are all exactly 5C lower than SpeedFan.

The temperature readings for my graphics card and four RAM sticks are identical however between SpeedFan and SIW.

After seeing this thread and also looking at an Intel document on the physical quadcore chip layout (the core centers are in-line and quite close to each other), it weems unlikely that this is a thermal paste/heat sink issue.

If anyone out there has more insight regarding the interpretation of these readings please let us know. I don't like the 60C reading considering that the Tcase (not core temp) max specification is 67C. It looks like instrumentation error so I'm not too worried though.

Reply to skypod1
reynod   02-22-2009 at 04:24:34 AM
- 0 +

If it works ok after regreasing and reseating the HSF then don't work about it.

I get at least a 4 degree differential between all 4 cores on my Q.

Check one of my X2's and it is about the same as yours - 5 degree difference at idle.

Obviously trying to lessen it by fiddling with the cooler / paste is a good idea, but if it isn't a problem running under load then don't worry.

It isn't like your watching the guages on the Challenger Shuttle ...

------------------------------ Great spirits have always encountered violent opposition from mediocre minds

 

Reply to reynod

against   07-27-2009 at 10:59:57 AM
- 0 +

I have an Intel E8500 Dual Core and the same problem, with two different coolers. The stock one and an Arctic Cooling ALPINE 7. With Core Temp, there are huge differences between cores, for example:

Idle
Core 0 - 50 C - 0%
Core 1 - 32 C (varying from 32 to 44 degrees) - 0%

Load
Core 0 - 58 C - 100%
Core 1 - 58 C - 100%

I've tested it using Prime 95 and the difference appears only in idle. As soon as it starts to work, temperatures even up, but at idle, there's a problem. Do you think you can help me?

Thnx.

Reply to against
randomizer   07-27-2009 at 11:50:28 AM
- 0 +

Your sensor is bottoming out or "sticking". It's not actually running at 50C at idle, the sensor simply doesn't respond below 50C. Nuff said in an old thread.

Reply to randomizer
Tom's Hardware > Forum > CPU & Components > CPUs > Core 1 10.C over core 0?
Go to:

There are 676 identified and unidentified users. To see the list of identified users, Click here.

Forum map
Bestofmedia Forum ©
2000-2009 Bestofmedia Group
Page generated in 0.578 seconds
Sponsored links
  • Ask the community now
  • Publish
Ad
Related Topics
See all relatives topics
They won a badge
Join us in greeting them