Re: [OctDev] complex error function

[Steven G. Johnson]

On 11/21/12 11:53 AM, Michael D. Godfrey wrote:
> By the way, the implementation of complex erf as defined for
> C99 is underway. The specification has been available for
> some time. Again, just something to keep in mind.
> In due course this will be the natural, maintenance free, source.

Two problems:

First, the C99 standard 
(www.open-std.org/jtc1/sc22/wg14/www/docs/n1124.pdf) does not define erf 
for complex arguments.  It only suggests the name cerf (etc.) as a 
possibility for future use; see section 7.26 "Future library 
directions".  Do you have any information on this being turned into an 
actual part of the standard?

Second, the standard only defines erf and erfc, and not the scaled error 
function, which can be computed accurately over a far larger domain.

Steven