Intel Core Temps

[Apoptosis]

I hope you guys find this to be a very interesting articleabout core temps. This article was really fun to write!

So the next time your buddy is bragging that his cooling solution on an ASUS board is better than the one on your ABIT board just laugh and say whatever! If he doesn't believe you show him this article and put him in his place! At the current time it is not possible to consider comparing temperatures between two different brand lines.

Let me know your feedback! If this article helped you out or just flat out confused you let me know! It's always good to have reader feedback to make my future articles better.

P.S. Anyone catch our 4GHz Prescott? hehe

[mcklain]

Finally an official review on temp differences!! I knew there was a difference between the p4c800-e and the IC7-Max3 and the cpu temps, but everywhere I read people were talking about 7-10degreese different but youre test really secured me. At home I have both those boards running a p4 2.4c. On my Asus p4c800-e my 2.4 is @ 275fsb and with the case open and an mcx478 rad with a panaflo high I was getting about 23 idle and about 28-30 with my case closed. Now when I received my IC7 max3 about 1 month ago, the temps were really disturbing me. I was running my other 2.4c @ 285fsb with a panaglo high on a sp-94 rad and was getting about 36 idle case open and about 40 idle case closed. At first I though those temps where high compared to my Asus board, even if I reduced the temps by 7 degrees, wich at that point I though was the correct difference between the boards. But youre review confirmed to me that my temps are extremely ok. If i substract 13 to 36 it gives me 23 wich is adentical to what I was getting the case open on my asus board!!

Thanks guys, finally someone did an article on this important fact!!

[Tread70]

Good article but that still raises one huge question? Which one is closer to the correct temp? If I was to venture a guess it would be Asus's because most other motherboard manu's have close to the same as them. In Abit's forum they had a thread about temps and they said they use the calculation formula that is given from the I/O vendor. So, who is closer to being correct? Would make another great article to venture into to see what the actual differences of temps using the same cpu in different mobo's to get an actual temp and compare it to the temp the mobo is calculating it to be.

Tread

[Apoptosis]

Tread,

We kind of did that with the northwood testing.

Below are the results with a 3.2C at idle.

ABIT: 36C
ASUS: 23C
INTEL: 27C

As you can tell not one of them is "close" to another. My bet is that the Intel board is close to being correct. I'll see what else I can dig up for everyone. This is a pretty in-depth subject and takes a ton of time to get in touch with the right people to find the right answers out!

Thanks for reading the article

[cold@ice]

@ Apoptosis,

very good work, thank you for this fine research.

i have three qustions:

-do you think it exists similar effects with AMD (64) CPUs and Boards?
-maybe you write an article for AMD also in the next time?

-and, it is important wich version of BIOS is used? or is this unimportant
in this case?

cold@ice

[Apoptosis]

I believe that this situation exists in both AMD and Intel based boards.

I am currently busy with a site redesign and testing some upcoming processors for an upcoming launch, but after the launch I will look into the AMD cpu temps.

The BIOS does play an important issue, but I played with BIOS 1014 and 1015 on the ASUS board and saw no difference in temps. I've updates the ASUS to 1016, but haven't had the time to look into it closer. (I don't see any difference as of yet)

[cold@ice]

@Apoptosis

thank you. I wait for your test and i hope that you can finish your work
soon as possible...

Some ideas to BIOS, because you say it is important:

It is maybe interesting what the author of the BIOS has change
in the BIOS code and why he has change something:

New orders/specs from AMD/INTEL or
the random factor is to big (or something else)...

if exists the possibility that somebody can disassemble the BIOS Code,
maybe it is possible to take a comparison between manufacturers and
the used hardware, the reason why its necessary to change something,
or whatever you find out... it is eventually helpful to understand, how
manufacturers work.

also, maybe we can see wich manufacturers works most correct - and how. But i think this is not simple...!

thanks,

cold@ice

[Apoptosis]

I tried to start looking at some AMD temps on an ABIT AN7 but quickly found out that MBM5 does not support that board and either does ABIT EQ. I used the uGuru software utility and it works, but wont so the high/low temps...unless I'm blind and am missing it.

sigh... 3 hours later I have no data, but a Barton 3200+ with a fresh install of XP on a 10k raptor... woopie

[mike]

To Legit Reviews testor

just a sugestion, try an external probe on the cpu, stick to edge of underside of heatsink next to cpu heat spreader. superflower makes a 4 probe sensor / fan controler with probes that could be stuck under the heatsink. I would calibrate it first with a knowen to be accuret heat source. The probes are on long wires so could be brought to the outside of the case for calibration.

[Apoptosis]

Mike,

While a valid point it's kinda not the point of the article. The article was trying to show that the "calculations" used to find the internal core temp varied between board makers. Thus rendering the temps found via the mobo/cpu sensors to be of limited use.

If we wanted to check to see what the "external surface" temp of the cpu is we can use a probe, but the value would be greater than 10C cooler than the hot spots in the core.