On 6 Jun 2004, at 22:02, Bob Smith wrote:
Okay.... I still would like to hear comments one way or the other as to why averaging readings is or isn't a good idea.
It is a good idea and as a concept underpins pretty much the whole of science and engineering (and woodwork: -measure twice, cut once <g>). If you measure it once, it may not tell you much, if you measure it twice and get wildly different answers you know your measuring system is no good or what you are measuring is changing with time. If you measure it lots of times you can start making statements about how accurate your measurement is. Taking an average of all the measurements is (normally) the best answer that the particular device can come up with.
When you are measuring a patch produced by a desktop printer, it is only approximately even, so it does not have a single answer as to what colour it is. When you measure the 'same' patch from a different print, you get a different answer. How different depends on the quality of the printer etc etc. An average of the two is a more 'accurate' answer than each answer individually.
(Yes I know the above is a simplification of a complicated subject <g>)
Matthew Ward
=============================================================== GO TO http://www.prodig.org for ~ GUIDELINES ~ un/SUBSCRIBING ~ ITEMS for SALE
