## 390 Reputation

11 years, 136 days

## @Mariusz Iwaniuk Hey and Thanks. The poi...

@Mariusz Iwaniuk Hey and Thanks. The point is, is it possible to reexpress Ei1 by Ei or somewhat different, in order to get rid of the I*Pi term? I'd like to have this function be represented by manifest real functions.

I have a different expression which is not yet simplified:

(-(exp(j*z))^2*Ei(1, j*z)+ln(-z)-ln(z)+Ei(1, -j*z))/(2*j*exp(j*z))

The problem is that, Ei(1,z) jumps by 2Pi when crossing the negative real axis.

In fact it seems that Ei(1,-j*z)=-Ei(j*z)-I*Pi so why is this not simplified?

## @Kitonum Hey thanks for this. Is it poss...

@Kitonum Hey thanks for this. Is it possible to use this in general when simplifying expressions or do I still in the end explicitly need to say

assuming R-sqrt(R^2+z^2)<0

?

## @Mariusz Iwaniuk  " Maple hav...

" Maple have some limitation about regarding the resolution of the case and falsehood."

Why is that? Why is it so difficult for it to realize that e.g. coulditbe(R-sqrt(R^2+z^2) < 0);#FAIL

is actually TRUE when assumptions are set?

Using with(MathematicalFunctions); and Assume() does not do any better or?

## @Mariusz Iwaniuk So in my case this is r...

@Mariusz Iwaniuk So in my case this is relevant:

convert(signum(0, x,1), piecewise);

#piecewise(x < 0, -1, 0 <= x, 1)

However as R-sqrt(R^2+z^2)  is obviously <0 then why is it not simplified as -1 ?

## @Carl Love Yes x>0....

@Carl Love Yes x>0.

## @gkokovidis ok thx.Assuming I have the f...

@gkokovidis ok thx.

Assuming I have the following expression

f:=exp(I*3*x-x/2)/(1+I)

Now when using the procedure from Preben gives me a rather weird result.

I therefore went on and removed the x terms

remove(has,f,x)

then I use

PolarForm(%)

and finally

combine(select(has,f,x)*'%')

However I still can not combine the result i.e. even when I double quote ''%'' it expands out again.

## @nm This seems awkward.. I actually have...

@nm This seems awkward.. I actually have other exponential functions and combining this number also in the exponential seems to simplify everything significantly.

## @Carl Love No I'm fine with it and I...

@Carl Love No I'm fine with it and I of course understand the reason behind it. Though when considering such sums which are only convergent within some domain the result I expected from maple evaluating it for me is always the analytic continuation. I just didn't expect it to be changed, because I thought everyone who is using maple knows the strictly speaking inequivalence here automatically.

## @Carl Love Well if you call df/dx a new ...

@Carl Love Well if you call df/dx a new function, then yes.

## @vv  What changed compared to Maple...

@vv  What changed compared to Maple17 where I do not need all this additional restrictions.

## @Christopher2222 Well thanks I guess ;)...

@Christopher2222 Well thanks I guess ;)

## @Christopher2222 I have an exponential f...

@Christopher2222 I have an exponential function with a prefactor and only the prefactor is supposed to become the big root...that is all the factors not in the root of the prefactor should go there.

## @Kitonum Yes, something like that....

@Kitonum Yes, something like that.

But I want to do it with a given expression. Not something i type in.

## @Kitonum Maybe my example was not the be...

@Kitonum Maybe my example was not the best.

ex:=sqrt(x)/2h*exp(-x^2/2*epsilon)

The prefactor should be fully in the sqrt() but the exponential should be left as it is...

Also I do not know where the exponential is so using op(*,ex) is also not really an option.

 First 6 7 8 9 10 11 12 Last Page 8 of 19
﻿