Different MB = better cooling?

[Dylruss]

Simple question that I haven't been able to make sense of......

I had a Soyo K7ADA mb (AliMagic chipset) and it was a good board, but seemed cheap to me, so I went out and got a Shuttle AK31a (Via KT266a chipset) board.

Here is the question:

I have the same hardware except the different mb's, but somehow my load temp and my system temp seem to be lower. So is it possible that a different mb could give you less heat on your cpu and system? Is it a stress to the system related issue, or something else? The shuttle has a vcore of 1.75 and the Soyo ran at 1.71, so not much difference there. This I do know, is that I get about 4 degrees less on the CPU, and 3 less for the system with the Shuttle.

Anyone know why?

[Demont]

Originally posted by Dylruss
Simple question that I haven't been able to make sense of......

I had a Soyo K7ADA mb (AliMagic chipset) and it was a good board, but seemed cheap to me, so I went out and got a Shuttle AK31a (Via KT266a chipset) board.

Here is the question:

I have the same hardware except the different mb's, but somehow my load temp and my system temp seem to be lower. So is it possible that a different mb could give you less heat on your cpu and system? Is it a stress to the system related issue, or something else? The shuttle has a vcore of 1.75 and the Soyo ran at 1.71, so not much difference there. This I do know, is that I get about 4 degrees less on the Shuttle.

it could be coincidence, maybe you mounted it a bit more even this time. or it could be that the board just reads the temps differently than your old one... that is the most likely.

[SkiFletch]

yeah, most likely that it just reads temps differently. also possible that the different layout of the mobo allows for different airflow through the case...

[Dylruss]

Well that could be true I guess, I am getting this from the bios itself, when I was checking temps. And as far as better airflow, it could be possible. I know the Soyo board was a little smaller then the Shuttle one, so I guess more surface area would allow more movement over it. I don't know, just thought that anyone would know if a different board really could run cooler for some reason.

[parkan]

I believe that is entirely possible. That's why the HSF tests are done with a CPU die simulator instead of a mobo.

[SkiFletch]

its highly doubtful that the different boards run cooler or hotter. it is however remotely possible that the different northbridges (ALi and VIA) put out different amounts of heat, but I highly doubt its anywhere near responsible for the range that you are reporting since most chipsets are VERY similar.

[rlemieux]

Most likely its just different readings. Some mobos read temps differently. The temps from the mobo are just a reference point. When I went from my P4TE to my TH7II my temps jumped by about 10C, just because of the onboard sensors. If you measure the temps with a probe you will notice they are actually close to the same, I know I did.

[Dylruss]

Yeah I will get a more accurate temp reader. I think that the Soyo fan was having some preblems also even though it was pretty new. And I am pretty sure that the Ali chipset is equivilent to the Via 266 chipset.

Thanks for the idea's guys, I was just curious if it were possible that one could generate more or not. But in the end I guess it must be some other reason like the different mb readings not the board itself.

[Tiger]

There are two possible causes;
1. The position and type of thermistor will give different readings.
2. You may or may not have recorded the vcore readings. Setting the vcore at say 1.75 may not necessarily mean the vcore is 1.75. An example of this is my two MSI mobos. The KT3 feeds 1.8 and my K7T Turbo feeds 1.86 at a bios setting of 1.75. This is a big difference and so consequently the K7T runs much warmer than the KT3.

[JasonKosi]

It's also possible that the CPU socket is in a different spot in relation to airflow, chipset, hot capacitors and FETs. Also, the position of the motherboard thermistor could affect temp reading.