As a very rough rule of thumb (with all sorts of caveats), 10 db equates to a perceived difference of 2:1, thus 10dB greater will sound twice as loud, and 10dB lower will sound half as loud. Similarly, a
3dB change will be detectable if switched quickly from one level to the other, but not detectable in a longer-term comparison test (using a chi-squared test for example). Bear in mind that the response of the ear is quasi-logarithmic, rather than linear.CRB
In message , The Natural Philosopher writes
1dB is reckoned to be about the smallest difference you can detect listening to a pure tone; 3dB is about the smallest difference for 'normal' sounds.All other things being equal, 43dB will sound noticeably quieter than
52dB.Cheers
Putting on my super-crypto-decoding hat:
Word 1 starts with a U. I think it's Use. Word 3 starts with a G. As we're talking of searching for info, it must be Google. Word 4 starts with a Fuh. Like Ferry, as in Gerry and his Pacemakers. But here I think it's another 3-letter job; specifically, "for". Word 6 then pops out as the four-letter word "sake".
Words 2 (or at least its unparticipled root) and 5 are spelt out with a single transposition error on T-shirts bought overpriced (or knockoffs at market stalls) by wannabe rebelious yoof. Yeah, go on, rebel: be just like all the other induhviduals making a personal fashion statement against global capatalism by buying up the image of a global brand...
Hey, Andy, can I have that job at GCHQ now? ;-)
"Dee" wrote in news:bmeee4$1nvm$ snipped-for-privacy@news.icl.se:
Don't worry about it, if everybody was a google expert the ng would have to shut up shop, unless we could find a few wannabe trolls to have some fun with - and I wouldn't have seen that handy list - I'll make a point to use earplugs on a rocket lauching pad next time
mike r
OMG. I start jabbering about subjective decibels, and of course someone from The Big Shed pops up, where they actually know this stuff.
Get me my job at HP back and I'll see about hollowing out your very own volcano
-- Die Gotterspammerung - Junkmail of the Gods
You probably wouldn't want to go back. HP has changed enormously in recent years since Carlie took over, but it was changing before that under Lew Platt.
PoP
Over 25 years ago I was in HP Grenoble on a training course. Dave Packard was in town holding some meetings around Europe. He did a presentation in the staff canteen - with John Young who was CEO prior to Lew Platt (showing my age now...).
Absolutely brilliant podium speakers, both of them. Highly motivating, easy to see how HP got to where they did. It was sad when Bill and Dave were no longer part of the company, things changed significantly at the sharp end of customer sales and support where I was based.
HP Labs in Bristol - been in there a few times on fleeting visits. Wonderful place. Pinewood was the other UK facility where it was real nice to work - drove past there yesterday, bit of a shame to see the For Sale hoardings outside - I remember the UK MD announcing in the Winnersh canteen that it was being opened.
PoP
I have my eye on that AEG too for the same reason.
Lets us know what's it's like if you purchase it?
Suzanne
PS Can I just point out that a baby can cry at 115dB! -somewhere between a pneumatic drill and a rock concert in front of the speakers. Hearing damage starts at 90dB.
Weighted to allow for ear's lack of sensitivity to low and high frequences.
Many thanks for all the replies. Off work yesterday but read them all this morning. I've learnt alot again. Oh how I wish I made more attention at school. Logarithims and noise levels we must have studied it but sure as hell have forgotten. Well as 43db is 10 times quieter than 52 then the AEG will be the one
AndyP
It's not exactly clear what "10 times" means... Almost certainly not that you could have 10 of them on at the same time and only making the same amount of noise as 1 of the louder variety. Fwiw (and there are some sound gurus lurking around here), I believe the average ear can only discern a change of about 3dB (maybe 2). 10dB is perceived as an approximate doubling in volume, iirc.
Nevertheless, the quieter machine will be noticeably quieter, but perhaps not throughout its full cycle. They probably use some averaging algorithm to quote results.
Think you should read the thread through again. ;-)
To the average human - not measuring instrument - a 10dB reduction results in a halving of the perceived sound level.
I'm fairly certain the origin of the decibel (or bel, of which it is a tenth) is that it is the minimum change in level that the average ear can determine in its most sensitive range. 3dB might well be a more practical amount, though, as an average across the audible spectrum.
I'd hope they're quoting the peak level, and it should be 'A' weighted to take into account the varying sensitivity of the ear to different frequencies.
Nah, the magnitude of a decibel is simply based on 1/10 of a log10 and, so, is mathematically determined. However, you may be getting confused with your story, as when calibrated for use as a noise measure, 0dB is set at a standard threshold of human hearing level by convention.
Christian.
Well, regardless of where you start the scale, the bel must be a 'quantity' based on something as any unit is.
Actually no - it's a "dimensionless" unit, 'cuz it expresses the ratio of one measurement [in this case of a sound level] to another measurement of the same thing. So the 'base of the unit' as you put is divided away to nuttin'. And hence the insistence of the pedants on saying "you can't give a value in decibels without also giving the reference value"; they're right in an anal sort of way, but for common level measurements there's a widely accepted reference level - for audible sound pressure 0dB is "person with good hearing (barely) can't hear it [but if it's made a bit louder they will]; for audio levels it's 0.775V into 600 ohms if I recall correctly. To express a wide range of ratios with smaller numbers, we then take the logarithm of the ratio, so that a difference in level of 1000 times is 3 bels (1000 being 10*10*10 - three tens), a difference in level of one-hundredth is -2 bels (1/100 being
1/(10*10) - two tens on the bottom). And in a final twist to confuse the unwary (maybe for marketing reasons? gawd knows) we use the decibel, tenth-of-a-bel, for common use; so a ratio of 100 gets called 20dB, a ratio of a million is 60dB, and a doubling is just about 3dB, since the base-10 log of 2 is about 0.3010 from memory (you can tell I went to school when log tables were still standard issue - but only just!).It's as if we measured speeds as a ratio to some Standard, say a British Standard Walking Speed of 4mph. Then a speed of 30mph would get called "7.5", a speed of 1mph would be called "0.25", and so on. If the Metrication Council then demanded we recalibrate the British Standard Walking Speed to be expressed as 6.437376 km/h, the 30mph = 48.28032 kp/h would still be called "7.5" (wot with
48.280832 being 7.5 times bigger'n 6.437376). The underlying arbirtrary unit has been divided out, leaving the "dimensionless" ratio.HTH, Stefek
Almost correct - I think? Except 0dB in audio is actually 0dBm which equates to 1mW into a resistive load, by convention accepted as 600R unless otherwise stated. 1mW into 600R is almost exactly 775mV.
Indeed. And with 'voltage' dBs, which will be the type used for quoting the loudness of appliances etc, 6dB is a doubling of the actual voltage measured. 3dB is a doubling of power, so applies to amplifiers, etc.
Not quite. A bel is a ratio, not a unit.
CRB
Why give it a name then if it is purely abstract?
FWIW, my dictionary says it's a unit for measuring sound.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.