[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Alternate/Unusual Primary Configurations

Original poster: "Barton B. Anderson" <bartb@xxxxxxxxxxxxxxxx>

Hi Matt,
Yes. We can only measure to instrument accuracy and technique. My statement was regarding basic equipment. For example, an instrument with say a .005% accuracy at some parameter and the sim matches spot on. At some point, the instrument cannot be trusted because it isn't designed to go further. However, if the general population of instruments average .01% at the same parameters, then the program is more accurate than the general population can measure. That's all I was referring to. 

Take care,
Bart

Tesla list wrote:


Original poster: Mddeming@xxxxxxx
In a message dated 4/13/07 2:24:42 A.M. Eastern Daylight Time, tesla@xxxxxxxxxx writes: 

<<<<snip>>>
Programs today can get
the coupling exact and better than you can measure. It's weird,
coupling for the longest time was one of those "must measure to be
even ballparked", but today, the programs are "actually" better than
the measurement (seriously).

Hi Bart,
It seems to me that if the simulation is more accurate than measurement, then that accuracy cannot be verified. If accuracy (of a model) is defined in terms of how well it matches reality, and experiment is the final arbiter of reality, and the model is superior to experimental measurement, then we have ourselves an epistemological conundrum. It may, in fact, be more accurate, but we can never know that, except by "faith". Perhaps "accurate to the limits of measurement" would be a better way to put it. 

Matt D.




----------
See what's free at <http://www.aol.com?ncid=AOLAOF00020000000503>AOL.com.