[Subject Prev][Subject Next][Thread Prev][Thread Next][Subject Index][Thread Index]

Re: "time complexity"

On Wed, 17 Jan 2001, Shridhar Daithankar wrote:
> I mean I know how important it is but I would rather prefer to compute in
> theoretically ( which is going to be inaccurate) rather than computing it by
> examples.

I guess thats the way its done. Computationally you can get the the
*time* not the time complexity. One can always try plotting "N" versus
"time" and fit some kind of a least square polynomial fit, but then there
are NP hard problems which will not fit those polynomials. Moreover big
O() notation disregards all but the highest index. So even if I am trying
to get the complexity by curve fitting, the real nature will come forth
only at big values of N.

> P.S. My greycells are already dead. Why would I do a job otherwise? :-))

 I thought they always died after the twelves, thats if they have lasted that
long to start with ;)

> Amarendra GODBOLE wrote:
> > On Wed, Jan 17, 2001, the greycells of Shridhar Daithankar expressed: