Re: Help! Integrate...
- To: mathgroup at smc.vnet.net
- Subject: [mg9021] Re: [mg9007] Help! Integrate...
- From: David Withoff <withoff>
- Date: Wed, 8 Oct 1997 00:05:19 -0400
- Sender: owner-wri-mathgroup at wolfram.com
> Hello again!
>
> Thanks to everyone who responded to my last post, and here's another:
>
> I just today upgraded to Mma 3.0, and immediately noticed one step
> backwards. Under 2.2.4 I gould give Mma the folowing integral
>
> Integrate[T''[x]Cos[x],{x,0,1}]
>
> and it would churn through the integration by parts to get a somewhat
> evaluated for of the expression. Now, under 3.0, it just stares at me and
> stupidly spits back exactly what I give it.
>
> 2.2.4 would always load its integration packages when it first
> encountered the integral, and 3.0 doesn't. Is there a way to load them
> myself, or is that functionality gone with the improvement?
>
> Thanks again!
>
> Greg Anderson
> dwarf at wam.umd.edu
Some incompletely integrated results, including the result for this
example, are no longer returned in Version 3.0. This was a difficult
choice, since these results are occasionally useful, but it is also
true that incompletely integrated results are often annoying, they
slow everything down, and they can lead to infinite loops. We got
a lot more requests to get rid of them than to add them. Hopefully
there is some other way to do what you want to do.
In Version 3.0 the integration packages are loaded automatically, and
they are loaded a *lot* faster than they were loaded in Version 2.2.
You don't need to do anything to load them.
Dave Withoff
Wolfram Research