Rainer Joswig wrote:
> In article
> <eb9c4b9a0705051636q1c75884cr24a18fbadf1ed550 at mail.gmail.com>,
> "Robert Dodier" <robert.dodier at gmail.com> wrote:
>
>> On 5/5/07, Rainer Joswig <joswig at lisp.de> wrote:
>>
>>>> Sorry, my mistake. I asked the wrong question. I meant to say,
>>>> What does (INTEGER-DECODE-FLOAT 5.2d20) return?
>>> MAXIMA 19 > (INTEGER-DECODE-FLOAT 5.2d20)
>>> 7934570312500000
>>> 16
>>> 1
>> OK, that seems like the correct result. At this point I give up.
>>
>> I'm still interested to know why 7.000000000000013E19 yields a
>> test failure compared to 7.0E19.
>>
>> Thanks for your help,
>> Robert
>
> It boils down to this:
>
> LispWorks 5
>
> CL-USER 6 > (expt 1.0d+20 2)
> 1.0000000000000016E40
>
> CL-USER 7 > (* 1.0d+20 1.0d+20)
> 1.0E40
>
> most Lisp systems I tried are giving the same result
> for both forms.
>
I'd file a bug report to LW. There's really no excuse these days for
this to be inaccurate.
Ray