Hi. I have a TI 83+ and a Ti 92 (not plus). I'm trying to make the 92 more useful for statistics, so I created a seemingly analogous normal cumulative distribution function on it:

normcdf(zscore1,zscore2)

Func

(comment) equation for normal cumulative distribution function

(sqrt(2*pi))^(-1)*integral(e^(-t^2/2),t,zscore1,zscore2)

EndFunc

This does give the right answers. However, it runs much slower than normalcdf() on the 83+, especially with large arguments. The 83+ responds rapidly even with one of the arguments is, say, 10^99.

Is the 83+ using a different algorithm that is much more efficient? Is it the integration that's really slow?

Thanks for any insight.