Robert Dodier wrote:
> On 4/14/07, Detlef Schmicker <d.schmicker at physik.de> wrote:
>
>
>> if I use a self defined function
>> dint(f,v,a,b):=subst(b,v,integrate(f,v))-subst(a,v,integrate(f,v));
>>
>> instead of integrate for the integral with bounds, no questions are ask,
>> and the result seems to be OK.
>>
>> Is there a reason, build in integral is working differently?
>>
>
> Maxima's algorithms for definite integrals are mostly different
> from the algorithms for indefinite integrals. Maxima might try
> to evaluate a definite integral via the fundamental theorem of
> calculus but usually it tries other methods first (if I recall correctly).
>
>
Yes, that's exactly right. Maxima tries to convert definite integrals
to contour integrals which can be evaluated using residues. But maxima
can also try the indefinite integral and substitute the limits. I don't
know why the indefinite integral is tried fairly late. Plus there's
some bug in the code where maxima gets the indefinite integral wrong (or
gets stuck), but you if do the indefinite integral yourself, maxima gets
the right answer.
Ray