Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.soft-sys.math.mathematica > #6497

Re: Error in FullSimplify

From Andrzej Kozlowski <akoz@mimuw.edu.pl>
Newsgroups comp.soft-sys.math.mathematica
Subject Re: Error in FullSimplify
Date 2011-11-17 12:03 +0000
Organization Steven M. Christensen and Associates, Inc and MathTensor, Inc.
Message-ID <ja2t6a$n4j$1@smc.vnet.net> (permalink)
References <201111171104.GAA22782@smc.vnet.net>

Show all headers | View raw


1. The error is no longer present in Mathematica 8:

In[17]:= FullSimplify[Rationalize[FullSimplify[-E^(x (a+I b))+E^(x (a-I b))+E^(x (-a+I b))-E^(x (-a-I b))/. a->3./. b->4.]]]

Out[17]= -4 I sin(4 x) sinh(3 x)

2. FullSimplify depends on such things as exact cancellations, so using it with approximate numbers is not a good idea. However, it has to be admitted that Mathematica is getting much better at dealing with such situations, and it is getting hard to find examples to illustrate the pitfalls.

Andrzej Kozlowski


On 17 Nov 2011, at 12:04, Philipp Wei=DFgraeber@smc.vnet.net wrote:

> Hello
>
> I have experienced some strange error in the FullSimplify-function in
> Mathematica 7.
>
> Input:
> (-E^((a+b I)x)+E^((a-b I)x)+E^((-a+b I)x)-E^((-a-b I)x))/.a->3 /.b->4 //
> FullSimplify
> (-E^((a+b I)x)+E^((a-b I)x)+E^((-a+b I)x)-E^((-a-b I)x))/.a->3. /.b->4.
> // FullSimplify
>
> Output:
> -4 I Sin[4 x] Sinh[3 x]
> -2 I Sin[4. x] Sinh[3. x]
>
> And the later result is wrong.
>
> What happens here, how can this be explained?
>
> CU
>
> Philipp
>

Back to comp.soft-sys.math.mathematica | Previous | Next | Find similar | Unroll thread


Thread

Re: Error in FullSimplify Andrzej Kozlowski <akoz@mimuw.edu.pl> - 2011-11-17 12:03 +0000

csiph-web