> Note that linspace defaults to interpolating 100 points; > this is pretty inefficient here, ...
> Also, there is the cis(x) function which does the same thing as exp(x*im). I know. Please not that this was meant as a test case only. And I wanted to the integration along the unit circle, not a triangle or square. Much better for testing... By the way: Sometimes I feel there are too many functions in the base system. Wouldn't it be more appropriate to put more functions of special areas into their own packages, as for instance Python does with the SciPy module (and others). On Friday, July 18, 2014 10:36:07 PM UTC+2, Steven G. Johnson wrote: > > > > On Friday, July 18, 2014 4:11:34 PM UTC-4, Peter Simon wrote: >> >> Also, there is the cis(x) function which does the same thing as >> exp(x*im). But I notice that it isn't (yet?) vectorized, >> > > Should be fixed now in git master. >
