Post by thread:PICLIST Digest - 23 Sep 1997 to 24 Sep 1997

>Of course, you can now pick N rather arbitarilly and keep >a "running average" by calculating: > avg = avg * (1 - 1/N) + sampN/N My two cents worth. This seems to be exactly what you get if you numerically simulate a simple RC lowpass filter with discrete timesteps. In this case samplN is the input voltage at time N and and avg would be the output (that is : the voltage over the capacitor). N is given by RC/dt where dt is the time step. This remark alows you think about the discrete filter in terms of familiar notions such as the 3dB point, etc... It could be worthwhile to look at discrete versions of higher order filters. There is something about this in Horowitz and Hill I believe. Michel