Not satisfied with i915GMm-HFS bundled cpu cooler?

chiccoweb · Post by **chiccoweb** » Sat Jul 16, 2005 1:07 pm

In these days I'm testing the Pentium M mainboard from Aopen i915GMm-HFS, a really great mainboard for an HTPC system. The only thing unsatisfied me is the satandard cpu cooler.

Surfing on forums I noted many users share with me the same feeling.
Well I have also a DFI 855GME with latest Microcool cooling systems (14dB), unfortunaty the Aopen mounting holes are in a different position so the NorthPole XE whisper doesen't fit.

Never mind a spent few minutes in my workshop and I modded the NorthPole clips for intel chipset.

Screws mounting systems is the same:

The clips need a little lifting:

OK now stay tuned for noise and cooling test results....

Mar. · Post by **Mar.** » Sat Jul 16, 2005 2:31 pm

The word is motherboard. That gender-neutral BS gets old...

Anyway, nice pics, I'm sure this will help someone out there.

chiccoweb · Post by **chiccoweb** » Sun Jul 17, 2005 4:24 am

I did some tests and I report here the results

The CPU

@ IDLE

@ FULL LOAD

I'm very glad to have temperatures really lower then stock cooler and a more noiseless solution.
Anyway temp reported by Aopen bios seem too high, what do you think?

Could be nice if i915GMm-HFS users post here cpu temp at idle and full load for comparison.

Mar. · Post by **Mar.** » Tue Jul 19, 2005 10:00 am

It idles at 46C? That's pretty high by comparison, but shouldn't pose any real threat... Sounds like my system.

chiccoweb · Post by **chiccoweb** » Mon Jul 25, 2005 7:19 am

IMHO the Aopen i915GMm-HFS doesn't read correctly the cpu temperature, take in consideration the same Dothan (1.86 GHz) on a DFI 855GME: the cpu temperature at full load is much lower.

Post by **Ralf Hutter** » Tue Jul 26, 2005 4:48 am

chiccoweb wrote:IMHO the Aopen i915GMm-HFS doesn't read correctly the cpu temperature, take in consideration the same Dothan (1.86 GHz) on a DFI 855GME: the cpu temperature at full load is much lower.

My testing shows that, using the same heatsink/fan combo on each board, the AOpen 855GMEm and 915GMm boards report the same CPU temp, at idle and at load. Based on that, and based on how hot the tiny little stock 915GMm HSF gets under load, I think the 915GMm is reporting the temps pretty accurately.

Specifics on this test can be found in the "System Testing" section of the AOpen 915GMm review.

fucoboy · Post by **fucoboy** » Sat Jul 30, 2005 3:21 am

So what would be the conclusion then? I am not a technician but logically seen the board seems to cause the CPU to get hotter than on other boards. Maybe this is because the voltage shown by the BIOS is not correct. But if the board would overvoltage the CPU I guess that this could be measured by someone. What other reason could cause that increased heat? Maybe we should call this a tropical board. If your CPU desires to get really hot let it visit the AOpen i915 board island. It will be satisfied....

Edit:
BTW, did anybody test the BORG HPC 479 cooler? This looks like a pretty fine solution. Currently it seems not be available. Maybe I will try out that one, if I can obtain it.

Post by **Ralf Hutter** » Sat Jul 30, 2005 5:15 am

fucoboy wrote:So what would be the conclusion then? I am not a technician but logically seen the board seems to cause the CPU to get hotter than on other boards. Maybe this is because the voltage shown by the BIOS is not correct. But if the board would overvoltage the CPU I guess that this could be measured by someone. What other reason could cause that increased heat? Maybe we should call this a tropical board. If your CPU desires to get really hot let it visit the AOpen i915 board island. It will be satisfied....

Sigh...

The board doesn't cause the CPU to "get hotter than other boards". The crappy little stock CPU cooler doesn't dissipate heat as well as other cooler, causing the CPU to run hotter on this board. Replacing the crappy default cooler with a decent cooler brings the CPU temps right into line with the temps reported by other boards. I don't know how to put it more plainly than that.

There's nothing wrong with the board or the voltage it supplies to the CPU, there's something wrong with the stock cooler. End of story.

Reachable · Post by **Reachable** » Sat Jul 30, 2005 7:02 am

Mar. wrote:The word is motherboard. That gender-neutral BS gets old...

...talk about irritable.....

This is hardly worth commenting on. Nevertheless, "mainboard" more accurately describes that piece of equipment than does "motherboard".

Below is an example where it makes linguistic sense:

A mothership is a vessel or aircraft that carries a smaller vessel or aircraft that operates independently from it. Examples include bombers converted to carry experimental aircraft to altitudes where they can conduct their research, or ships that carry small submarines to an area of ocean to be explored. The mothership may also recover the smaller craft, or may go its own way after releasing it.

en.wikipedia.org

It seems like the guys who made up this kind of terminology were being cowboys who liked to use colorful or harsh terms. Like "master" and "slave" drives, which really doesn't describe the relationship between the two drives.

Sorry to veer from the thread. Commendations to the initial poster for being industrious enough to mod his equipment and test it.

swivelguy2 · Post by **swivelguy2** » Sat Jul 30, 2005 7:07 am

Let's try to keep embedded images down to 640x480, as per the rules.

The 800-pixel wide images are stretching it, but probably ok in my book. However, the 1024-pixel wide desktop image is ceratinly a frame-breaker for many people.