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

Which is more accurate RealTemp 3.40 or CPUID Hardware Monitor 1.15?

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

Farwalker

Registered
Joined
Aug 13, 2005
Location
Gainesville, GA
Since heat can be a serious issue in overclocking a CPU, I would like to know which of the two applications is more accurate.

Heat generated by my CPU is now my main focus.
I am cooling with air Thermalright Ultra-120 eXtreme Rev C.
Room temps are around 73/23 - 76F/24C.
Of course my cooling does much better when I open the window at night and let 50F/10C air flow into the computer front fans.

So may question is:
Is RealTemp 3.40 or CPUID Hardware Monitor 1.15 more accurate.

RealTemp 3.40 consistently reports temperatures of 2 - 3 degrees Centigrade higher than CPUID Hardware Monitor on all cores. Those few degrees could lead to early CPU death.
 
I started typing a response, but after a paragraph realized that everything I was noting about Tcase, Tjunction, Tj Max, etc. (in regards to temp differences w/ different software like Real Temp, Core Temp, SpeedFan, etc., including calibration) was already included in both the 'Intel C2D/Q/X Thermal Designs Explained' sticky at the top of this forum, and the 'Core i7 and Core 2 Temperature Guide' over at Tom's...

Intel C2D/Q/X Thermal Designs Explained
http://www.ocforums.com/forumdisplay.php?f=2

Core i7 and Core 2 Temperature Guide
http://www.tomshardware.com/forum/221745-29-sticky-core-core-temperature-guide
 
TJMax value for Nehalem/Lynnfied is written in MSR so both programs should read same values.
Personally as a beta tester for RealTemp I trust RealTemp. :D
 
If you are worried about 2-3c making the difference you are pushing it too hard. My ambient tems fluctuate that much in a dayas my apartment faces west and I get the evening sun blaring in.
 
All programs should be reading the same value from the same sensor. Any of them is no more or less accurate than another. Precision (the coded TjMax, and therefore displayed temperature) is not accuracy :) There's no "magic voodoo" involved in getting the temp. Every program with the same coded TjMax is going to report the same temperature. Pick whatever program has a UI that appeals most to you.

TjMax keeps making me think "cheap-crap department store"...
 
In theory, all software should read TJMax from a register within the newer Core i7/i5 CPUs and should read the exact same data from the same temperature register in the CPU but for some reason, there are still differences.

I know RealTemp is able to do this correctly and I know Core Temp usually reports the exact same thing but there are other programs that continue to only read temperature data from 4 threads instead of from 4 cores. When a program does that, it's usually only reporting the temperature from 2 of your 4 cores.

As for CPUID Hardware Monitor, I have no idea what it's doing. I frequently see differences like you are seeing of a few degrees and I don't know why. The method is well documented and if two programs like RealTemp and Core Temp are able to follow this method then I don't know why other programs aren't doing the same thing.
 
RealTemp is what I prefer for CPU temps. Sometimes I use CPUID to get other system temps.

And you've got the author of RealTemp posting in this very thread. How's that for support!
 
Back