[time-nuts] OT: NZ Christchurch member
jimlux at earthlink.net
Fri Feb 25 16:09:12 UTC 2011
On 2/25/11 7:55 AM, Poul-Henning Kamp wrote:
> In message<AANLkTikVLp0FMTrm8waFFAD9voHXc8oEo2ZJiCpniQKQ at mail.gmail.com>, "Wil
> liam H. Fite" writes:
>> Me: You're saying that the Richter is a poor predictor of surface
> For damage assement you really need a vector-version of richter,
> vertical does a lot more damage than horizontal on average.
Yes.. I doubt anyone still uses the torsion seismometer Richter used,
although more modern scales (moment magnitude, etc.) still relate back
(e.g. they set the calibration to match for some notional set of events)..
That way, people have an idea... A Magnitude 3 earthquake within a few
tens of km of me will be noticeable, if it's quiet. A magnitude 4 will
be very noticeable, and a 5 will be exciting. A 6 will wake you up in
the middle of the night. I'd compare it to something like Mohs
hardness, except actually with a quantitative basis. (People who work
with material properties like hardness use other scales anyway)
It's a "roughly quantitative" measure of energy release, in the same
sense that kilotons are for explosions. It's like that whole "cup of
gasoline: dynamite" comparison.. it's the rate of energy (e.g. power)
that creates the qualitative difference between running my camping stove
We do the same thing in time-nuttery.. we use log scales to talk about
performance.. dBc/Hz for phase noise, and really, just the exponent to
talk about ADEV. (nobody gets excited about the difference between
1.1E-13 and 1.5E-13... but the difference between 1E-11 and 1E-15 is
worth talking about)
Maybe we should start promulgating dBallan?
And maybe get an SI unit... The "Allan", although since the fractional
frequency error is dimensionless....
More information about the time-nuts