Comments and discussion on Valentin's 4th "Then and Now" - Area

[Albert Chan]

(01-21-2023 04:16 PM)J-F Garnier Wrote:  I had the idea that it may be possible to write an adaptive integration code on the 71B by first analysing the integrand behaviour, actually I'm surprised not to find anything like that in the forum or various 71B code archives (I didn't do a extensive search though). Maybe is it just too complex, or just with no real usage.

I did Adaptive Simpson HP71B code a few months ago.
With that loaded in memory, we update function, limits, and eps.

>10 X1=0 @ X3=2.82740261413 @ E=10^(-8)
>50 SUB F(X,Y) @ S=SQR(-LN(X*X/30.07+EXP(-SIN(X))))
>60 Y=SGN(X+S)*ABS(X+S)^(1/3)-SGN(X-S)*ABS(X-S)^(1/3) @ STOP
>run
 2.07662636771      8.96 ! @200x, about 30 minutes for HP71B

SIMPSON can't see user defined function, so I coded with slower CALL F(X,Y)

Based on timings alone, I guessed F(X,Y) called about 3000 times.
I don't know how to count F(X,Y) calls. Variables inside F turned local. Any ideas?