[CUBE] Fan noise (was:Cube upgrades vs. new Powermac)

Joseph B. Gurman gurman at gsfc.nasa.gov
Sun Apr 25 07:59:14 PDT 2004


     atoa at krak.net wrote:

>You may be correct as factual matter, that the G5's are quieter than the
>G4's, but your logic escapes me.
>
>My G4, with two fans, isn't quieter than my IICI, which has only one.
>
>There's a lot more to noise than number of fans.

     Historically knowledgeable and logically correct. There is more 
to noise than the number of fans. Fan design and fan speed are part 
of it. Does the fan have a sharp edge? Is it so small it needs to 
rotate faster to move a given volume air per unit time than a larger 
fan could at lower rpm's? Does it rotate so fast that turbulence 
plays a major role in sound output? Are the bearings of such poor 
design that they start causing noise after a few (months, years) of 
use? And perhaps most relevant to this comparison, are they employed 
in a variable-speed design with feedback from temperature sensor(s)?

     The G5 is divided into three, physically separate cooling zones: 
memory and CPU('s), PCI/AGP cage, and optical/hard drives. In each 
one there are exhaust fans controlled by OS software that read zone 
temperature sensors. In the memory/CPU zone, there are intake fans in 
front of the CPU cooling stacks as well, also variable speed. Because 
the OS controls all this, the fans ramp up to full speed at boot (and 
that's not quieter than a G4) but become nearly inaudible in a room 
of reasonable temperature once the OS takes over. If a disk gets 
really active, you can hear those fans go on briefly, but most of the 
time, it's the disk drives (sadly) that you hear. If your workspace 
is much hotter than ~ 21 C, YMMV.

     I have an MDD G4 at home, one of the original, "wind tunnel" 
models. When it got warm in the room, it got noisy. With Apple's 
$19.95 "free" replacement power supply and fan kit, it still gets 
noisier than when cool, just not as noisy.

     And of course, how much heat there is to remove is the most 
important factor. The IIci had a 25 MHz 68030. Not exactly a barn 
burner in comparison to the 970, or even the Cube's original 
7400/7410 chip.

				Joe Gurman
-- 
"I love deadlines. I love the whooshing sound they make as they go by."
                                                             - Douglas 
Adams, 1952 - 2001

Joseph B. Gurman, NASA Goddard Space Flight Center, Solar Physics
Branch, Greenbelt MD 20771 USA



More information about the Cube mailing list