[plug] Hot and bothered CPU hankers for cool breeze

[Bernd Felsche]

On Tue, Dec 16, 2003 at 09:50:21PM +0800, James Devenish wrote:
> In message <20031216124328.GA8482 at erdos.home>
> on Tue, Dec 16, 2003 at 08:43:28PM +0800, Cameron Patrick wrote:
> > On Tue, Dec 16, 2003 at 08:40:54PM +0800, Derek Fountain wrote:
> > | I'm intrigued! Why do 400 boxes not sound 400 times louder than 1 box?
> > My understanding is that a 10dB increase corresponds to ~2x louder,

> Hmm...are we talking about intesity or power? In the case of intensity,

If dbA is mentioned, then it's sound pressure level with respect to
20 micropascals at 1000Hz, the "least discernable" sound by the
human ear.

3dBA does correspond to a doubling.

256 machines would be 8 * 3 dBA, i.e. 24 dBA louder if equi-distant
from the observer. 400 machines would be around 26 dBA louder.

If each machine puts out about 56dBA at 1 metre, then the chorus of
400 would be in excess of 80dBA; were they all at 1 metre from the
observer. A work environment where the subject is continuously
exposed to noise levels in excess of 75dBA will probably induce
premature hearing loss... unless they otherwise lead a very quiet
life.

Fan noise should be close to white noise; fans that "whine" are both
annoying and have a peak sound pressure level at distinct
frequencies; which can accelerate hearing loss as well, even at
"safe" levels.

To be reasonably fair to yourself, you should consider wearing
hearing protection if you're exposed to noise levels about 75dBA for
more than 8 hours a day. Workplace regulations, IIRC, require it
above 85dBA, based on an 8-hour exposure average.

> ~3dB corresponds to 2x more intense and 10dB corresponds to 10x more
> intense. (Power presumably involves the square of intensity, so it would
> be typical to calculate 20log(f(P)), as opposed to 10log(f(I)).) Of
> course, talking about power might be more practical since it relates
> readily to energy. But I think "louder" would be a perceptual
> phenomenon and we'd have to ask the audiologists.

> In message <20031216124752.GD7965 at erdos.home>
> on Tue, Dec 16, 2003 at 08:47:52PM +0800, Cameron Patrick wrote:
> > And from this we deduce that either I don't have a clue what I'm talking
> > about, or James slept through his Physics classes :-)

> I'm not sure that the 'semantics' of aural decibels are well defined for
> anything other than 0dB. That is, I have no intuitive reason to think
> that we can relate increases in intensity or power to perceptions of
> noise in a proportional fashion.

They are well-defined. Hence A-weighting for low to medium sound
pressure levels and C-weighting for loud sounds. Other weightings
exist, such as B, D, B+C, E and SI.

-- 
/"\ Bernd Felsche - Innovative Reckoning, Perth, Western Australia
\ /  ASCII ribbon campaign | I'm a .signature virus!
 X   against HTML mail     | Copy me into your ~/.signature
/ \  and postings          | to help me spread!