On 5/7/10 6:14 PM, Dieter Kaiser wrote:
> Am Freitag, den 07.05.2010, 17:53 -0400 schrieb Raymond Toy:
>
>
>> Hmm. I know there's a problem with this test with ccl, but not sbcl.
>> ccl fails because (expt 2 #c(-2d0 -1d0)) is not the same as (expt 2d0
>> #c(-2d0 -1d0)). Perhaps sbcl fails for the same reason? I don't have
>> sbcl currently installed.
>>
> I get the following results with SBCL:
>
> (%i1) :lisp (expt 2 #c(-2d0 -1d0))
> #C(0.19230972430417587 -0.15974031883619208)
>
> (%i1) :lisp (expt 2d0 #c(-2d0 -1d0))
> #C(0.19230972534099303 -0.1597403190784087)
>
>
That's what ccl produces too. cmucl and clisp produces the latter
answer for both.
I can't figure out from the spec what the right answer is, but I think
the latter is the correct answer.
Ray