jranger

Distinguished
Nov 4, 2010
32
0
18,530
using an intel q660 g0
software is cpuid heardware monitor, temps indicated are about 53*C
real temp 360 is telling me avg 24*C with a tj max set at 71....need to know which is right before i try anything that might blow up my cpu.... reason i may oc this one is because i am upgrading to a new core set of components to increase the effectiveness of my audio workstation, and trying to keep things cool without creating a noise machine..
any help would be greatly appreciated

J
 

jranger

Distinguished
Nov 4, 2010
32
0
18,530
I have two opposing temps from both of those software packages, the question is.... which one is right..... it would seem to me that the ambient or load factor would be somewhat irrelavent given the question has to do with two samples that are at odds at the same time, making things like load etc... irrelavent to the question.... i do understand that if i was trying to use one package and came up with different temps that this would all come into play.... fact is, 2 answers so different can't be right.... need to know which one is.. say for arguments sake..... ambient 26*C load factor 50% to 70%...
am i not setting up tj max properly in realtemp360,. and since cpuid hardware mopnitor doesn't have an adjuster for tj max... i am inclined to believer cpuid's temp since the motherboard bios gives me a similar temp at idle... heatsink is a thermalright tower 112, tdp 150w ..... so that is good and has been resat a few times in the past 4 years about... so the temps in the various flavours, seems to be consistent over time....

part of the reason i am asking is because i am planning on oc'ing the thing a bit, but seeing as this unit will be 100% loaded when it is being used i need to know what temp i am going to have to figure on, air cooled pref no fan on system at all, so.... this temp issue has to be right, or i'm goingto cook the cpu... and i don't want to do that... could do that without asking the right temp.... come to think of it.. anything over 90*C should be sufficient to cause damage.... but not asking that.... thanks


J
 
How do you know that the load on the CPU was the same in both cases? did you run the programs at the same time?

HWmonitor and Real Temp both display core temps... is that what you are looking at? And at the same time? like this?

tempo.jpg




If so then grab Core temp and compare to the other two, throw out the one that is wrong.
 
I'll try to explain this once more.

If you measure temps with one program, then measure temps with another program, you have no way of knowing if the load was identical. In fact it likely wasn't.

Even if you DO run the temp programs simultaneously, you may get slightly different temps as they have different poll frequencies and will sample at different times.

How can you know the load is identical in both cases? The answer is to run all 4 cores at maximum 100% load.
That gives you a hard, scientific number to work with when you subtract your ambient temp.

I gave you a simple solution. I sad to run Core Temp and compare, then toss out the reading that didn't match. You didn't do that, for some reason.

OK, so I did the work for you.

I googled up the actual Tjmax of your CPU, a Q6600 with G0 stepping
http://www.tomshardware.co.uk/intel-dts-specs,news-29460.html

While we are at it here is some further reading on the subject
http://www.hardwaresecrets.com/article/Maximum-CPU-Temperature/143/1

This is why HWMonitor doesn't allow you to mess with Tjmax btw. All the info is available via the CPU ID, so any program can auto-config the right settings... there is never a need to adjust Tjmax on newer hardware, that I have seen.