On Thu, Aug 12, 2010 at 12:29 PM, Olin Lathrop wrote: > Mark Rages wrote: > >> =A0FILTW <-- FILTW + FF(NEW*M - FILTW), FILT=3DFILTW/M >> >> M needs to be bigger than 2**N. =A0Practically, M should also be a power >> of two. > > Which is just a round about way of saying "fixed point". > No, fixed point is a red herring, equivalent to a units change. What I am saying is that right shifting by 1/(2**N) loses N bits of precision, and this can be compensated for by adding at least N bits to the accumulator. Regards, Mark markrages@gmail --=20 Mark Rages, Engineer Midwest Telecine LLC markrages@midwesttelecine.com --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .