Hello,

I'm running Openoffice 2.0.1-4bpo1, the backported version on debian
sarge, on a 1.40GHz Pentium M laptop with 1GB ram.

I have a simple spreadsheet consisting of three columns

   A         B       C
  action    change    running total
  -------------------------------
  get paid    +100          100
  buy apples    -8           92
  ...

where the running total in row n is computed automatically by
SUM(B1:Bn).  I'm finding that Calc is sluggish when I have about one
hundred rows.  For example, inserting a blank row in the middle takes
several seconds.

I tried a different style of running total where each cell is computed
recursively as something like Bn+C(n-1).  The performance is no
better.

It's clear that Calc is getting stuck doing quadratic work (O(n^2)) in
the number of rows. Is there a different way I should write the
formula to get better performance, since the problem is clearly linear
time.

By the way, this appears to a performance regression compared with
Openoffice 1.1.3, though I haven't actually measured it.

Many thanks for any suggestions!

-James Leifer

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to