Have done this on fairly small samples and it proves true what logically
(according to basic statistics) should be true. Over the longer haul the
variation from average gets smaller for larger weapons and thus they
outdamage the faster weapons if they have a better ratio. Unfortunately the
time needed to flatten tham out to average is longer than a few battles so
those battles which are below average willbe significantly below average and
you will die. The higher ratio always wins out in the long run, but if you
die due to fluctuation before that 'long run' can occurr does it really
matter?

I proved this by turning log on and off before and after fights and
recording them seperately. This allows one to see how greatly a wepon with
a delay of 50 (bloodforge hammer) differs from that of delay 28 (Shining
Star of Light). The bloodforge had a total over 5 fights that was higher
damage/time but it had 2 that were below the Shining Star despite having a
far better ratio. The Shining Star stayed very much the same over all 5
fights. Bloodforge 27/50 Shining Star 13/28. Obviously I did this test
with my cleric since both are weapons a bard cannot use.

Xabasbus via Kitasi

One thing you may want to do, that I haven't done yet. Get one of
the log parsing programs. They show damage/time, as well as %hit, and
several other stats.. Try several different combinations of weapons, and
find which seems to be the best total damage/time

Bien -- Karana