speedup help

[Bill Wood]

  . . .
> This code seems to compute 'bernoulli 82' in less then a second, in
> Hugs (on a 2GHz Pentium IV).

Just a note:  I compiled and ran Oleg's and mine for comparison.  The
two seem to be of the same complexity, with Oleg's a little faster
(modulo my using wall time; see previous msg.)

Oleg's blew the heap at 847; mine valiantly struggled on 'til it blew
the heap at 1910.  I must be doing something right, since I'm carrying
around the list of all numbers from B_0 through B_n, while Oleg cleverly
avoids that.  I was also surprised to see Oleg's blow the heap at an
*odd* value -- I thought he skipped those.

 -- Bill Wood
    wtwjek@winternet.com