[r-t] Predicting search size (was FKM)

[Don Morrison]

Mark Davies wrote:
> The (inaccurate, but generally useful) % complete measure in SMC32 

How's that work then? Is it comparing nodes visited versus some estimate of 
nodes that will need to be visited? If so, how do you construct the estimate? If 
not, what's it doing instead?



-- 
Don Morrison <dfm at ringing.org>
"The nice thing about standards is that there are so many of them
to choose from."                               -- Andrew S Tannenbaum