Bandwidth of partials

Interesting discussion on the bandwidth of partials in the PW forum. Starts off as a discussion of the relative value of tuning by M3's versus M4's and M5's, but that leads into a discussion (starting early but warming up on page 3) of how the bandwidth of a partial varies with sustain and possibly amplitude. Note that the discussion is about the bandwidth of individual partials, not all of the partials formed on a string. The subject comes and goes over the course of the thread:

http://www.pianoworld.com/forum/ubbthre...279/1.html

Makes me want to be able to experiment with this phenomenon. "The bandwidth of a partial will be about the inverse of the sustain time," according to DoelKees. As a feature? A bit much to deal with, perhaps. I do wonder about the difference it would make in the tone of specific registers. At the very least, it helps me to understand the furry, brief sound of the bass in most spinets and many console pianos. Seems to be a function of string length?

EDIT: I should have explained that DoelKee's view makes sense, in that more energy would be used by a wider bandwith--more portions of the string are moving, if more freqs are occurring, so that the sustain would be more brief, since the energy is used up faster.