Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: uncertainies, PROGRAMMING




On Thu, 12 Mar 1998 20:57:11 -0400 (EDT) LUDWIK KOWALSKI
<KOWALSKIL@alpha.montclair.edu> wrote:

Tim Folkerts showed that his spreadsheet results for 1600 calculations
are essentially the same as those from my 500,000 calculations with
TRUE BASIC. Then he adds: " .. it takes about the same time either
way".

Does it mean the same processing time *per case* or the same *for 1600
cases with spread sheet as for 500,000 cases with TB*?

I know this is not important in the context of our discussion. But I
would like to know the approximate time of your calculations, Tim, and
the speed of your machine. My expectation was that the spread sheet
approach is much much slower.
Ludwik Kowalski


What I actually meant was that the _programming_ times were
probably similar. If I were doing this a million times for a real
project, I would certainly use a program rather than a spreadsheet.

As you point out, for one calculation, the time to actually run the
progam is inconsequential - the main issue is the programmers time.
But for the record, I reran the spreadsheet on (a 180 MHz PowerMac) and
~1 second to redo the calculations after changing the width
~3 seconds to produce the distribution

(of course, it took much longer to move the mouse than to do the actual
calculations :)

You could always set up a macro to do both (and draw the graph) with a
single keystroke. My guess is that it would run in ~10 seconds. (But
at the moment I don't really feel like trying the 10 minutes worth of
programming.)






--- Tim Folkerts


********************************************************
Timothy J. Folkerts Tim.Folkerts@valpo.edu
Dept. of Physics & Astronomy 219-464-6634
Valparaiso University
Valparaiso, IN 46383