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


Groups > comp.soft-sys.math.mathematica > #3194 > unrolled thread

Re: Why can't FullSimplify give more uniform output?

Started byBob Hanlon <hanlonr@cox.net>
First post2011-06-20 12:05 +0000
Last post2011-06-20 23:37 +0000
Articles 2 — 2 participants

Back to article view | Back to comp.soft-sys.math.mathematica


Contents

  Re: Why can't FullSimplify give more uniform output? Bob Hanlon <hanlonr@cox.net> - 2011-06-20 12:05 +0000
    Re: Why can't FullSimplify give more uniform output? Jacare Omoplata <walkeystalkey@gmail.com> - 2011-06-20 23:37 +0000

#3194 — Re: Why can't FullSimplify give more uniform output?

FromBob Hanlon <hanlonr@cox.net>
Date2011-06-20 12:05 +0000
SubjectRe: Why can't FullSimplify give more uniform output?
Message-ID<itnd2c$7hj$1@smc.vnet.net>
Simplify and FullSimplify provide the simplest form that they find within their time constraints. Given different starting forms for an expression, there is no guarantee that they will arrive at identical forms.

$Assumptions = {u > 0, c > u};

T1 = (t1 - ((u x1)/c^2))/Sqrt[1 - (u^2/c^2)];

T2 = (t2 - ((u x2)/c^2))/Sqrt[1 - (u^2/c^2)];

dT = T2 - T1;

expr1 = FullSimplify[dT]

(c^2*(-t1 + t2) + u*(x1 - x2))/
   (c*Sqrt[(c - u)*(c + u)])

s = (t2 - t1 - ((u/c^2)*(x2 - x1)))/Sqrt[1 - ((u^2)/(c^2))];

s == dT // Simplify

True

expr2 = FullSimplify[s]

(c*(-t1 + t2 + (u*(x1 - x2))/c^2))/
   Sqrt[(c - u)*(c + u)]

expr3 = FullSimplify[Expand[s]]

(c^2*(-t1 + t2) + u*(x1 - x2))/
   (c*Sqrt[(c - u)*(c + u)])

expr3 is identical to expr1

expr1 === expr3

True


Bob Hanlon

---- Jacare Omoplata <walkeystalkey@gmail.com> wrote: 

=============
Hello,

I ask this question just out of curiosity.

Below is my input and output. The expressions "s" and "dT" are equal,
as can be seen by Out[17]. But the output from FullSimplify is
different for those two as can be seen by Out[12] and Out[15]. Why
can't they be simplified to expressions that look alike?

Thanks.

(PS. Also please tell me if my code is inefficient. I'm still learning
Mathematica)

In[1]:= Element[{x1, x2, t1, t2, u, c}, Reals]

Out[1]= (x1 | x2 | t1 | t2 | u | c) \[Element] Reals

In[7]:= $Assumptions = {u > 0, c > u}

Out[7]= {u > 0, c > u}

In[9]:= T1 = (t1 - ((u x1)/c^2))/Sqrt[1 - (u^2/c^2)]

Out[9]= (t1 - (u x1)/c^2)/Sqrt[1 - u^2/c^2]

In[10]:= T2 = (t2 - ((u x2)/c^2))/Sqrt[1 - (u^2/c^2)]

Out[10]= (t2 - (u x2)/c^2)/Sqrt[1 - u^2/c^2]

In[11]:= dT = T2 - T1

Out[11]= -((t1 - (u x1)/c^2)/Sqrt[1 - u^2/c^2]) + (
 t2 - (u x2)/c^2)/Sqrt[1 - u^2/c^2]

In[12]:= FullSimplify[dT]

Out[12]= (c^2 (-t1 + t2) + u (x1 - x2))/(c Sqrt[(c - u) (c + u)])

In[14]:= s = (t2 - t1 - ((u/c^2)*(x2 - x1)))/Sqrt[1 - ((u^2)/(c^2))]

Out[14]= (-t1 + t2 - (u (-x1 + x2))/c^2)/Sqrt[1 - u^2/c^2]

In[15]:= FullSimplify[s]

Out[15]= (c (-t1 + t2 + (u (x1 - x2))/c^2))/Sqrt[(c - u) (c + u)]

In[16]:= s - dT

Out[16]= (t1 - (u x1)/c^2)/Sqrt[1 - u^2/c^2] - (t2 - (u x2)/c^2)/Sqrt[
 1 - u^2/c^2] + (-t1 + t2 - (u (-x1 + x2))/c^2)/Sqrt[1 - u^2/c^2]

In[17]:= FullSimplify[s - dT]

Out[17]= 0

[toc] | [next] | [standalone]


#3197

FromJacare Omoplata <walkeystalkey@gmail.com>
Date2011-06-20 23:37 +0000
Message-ID<itolj6$eq9$1@smc.vnet.net>
In reply to#3194
Thanks for the answer.

On Jun 20, 8:05 am, Bob Hanlon <hanl...@cox.net> wrote:
> Simplify and FullSimplify provide the simplest form that they find within
 their time constraints. Given different starting forms for an expression, 
there is no guarantee that they will arrive at identical forms.
>
> $Assumptions = {u > 0, c > u};
>
> T1 = (t1 - ((u x1)/c^2))/Sqrt[1 - (u^2/c^2)];
>
> T2 = (t2 - ((u x2)/c^2))/Sqrt[1 - (u^2/c^2)];
>
> dT = T2 - T1;
>
> expr1 = FullSimplify[dT]
>
> (c^2*(-t1 + t2) + u*(x1 - x2))/
>    (c*Sqrt[(c - u)*(c + u)])
>
> s = (t2 - t1 - ((u/c^2)*(x2 - x1)))/Sqrt[1 - ((u^2)/(c^2))];
>
> s == dT // Simplify
>
> True
>
> expr2 = FullSimplify[s]
>
> (c*(-t1 + t2 + (u*(x1 - x2))/c^2))/
>    Sqrt[(c - u)*(c + u)]
>
> expr3 = FullSimplify[Expand[s]]
>
> (c^2*(-t1 + t2) + u*(x1 - x2))/
>    (c*Sqrt[(c - u)*(c + u)])
>
> expr3 is identical to expr1
>
> expr1 === expr3
>
> True
>
> Bob Hanlon
>
> ---- Jacare Omoplata <walkeystal...@gmail.com> wrote:
>
> =============
> Hello,
>
> I ask this question just out of curiosity.
>
> Below is my input and output. The expressions "s" and "dT" are equal,
> as can be seen by Out[17]. But the output from FullSimplify is
> different for those two as can be seen by Out[12] and Out[15]. Why
> can't they be simplified to expressions that look alike?
>
> Thanks.
>
> (PS. Also please tell me if my code is inefficient. I'm still learning
> Mathematica)
>
> In[1]:= Element[{x1, x2, t1, t2, u, c}, Reals]
>
> Out[1]= (x1 | x2 | t1 | t2 | u | c) \[Element] Reals
>
> In[7]:= $Assumptions = {u > 0, c > u}
>
> Out[7]= {u > 0, c > u}
>
> In[9]:= T1 = (t1 - ((u x1)/c^2))/Sqrt[1 - (u^2/c^2)]
>
> Out[9]= (t1 - (u x1)/c^2)/Sqrt[1 - u^2/c^2]
>
> In[10]:= T2 = (t2 - ((u x2)/c^2))/Sqrt[1 - (u^2/c^2)]
>
> Out[10]= (t2 - (u x2)/c^2)/Sqrt[1 - u^2/c^2]
>
> In[11]:= dT = T2 - T1
>
> Out[11]= -((t1 - (u x1)/c^2)/Sqrt[1 - u^2/c^2]) + (
>  t2 - (u x2)/c^2)/Sqrt[1 - u^2/c^2]
>
> In[12]:= FullSimplify[dT]
>
> Out[12]= (c^2 (-t1 + t2) + u (x1 - x2))/(c Sqrt[(c - u) (c + u)])
>
> In[14]:= s = (t2 - t1 - ((u/c^2)*(x2 - x1)))/Sqrt[1 - ((u^2)/(c^2))]
>
> Out[14]= (-t1 + t2 - (u (-x1 + x2))/c^2)/Sqrt[1 - u^2/c^2]
>
> In[15]:= FullSimplify[s]
>
> Out[15]= (c (-t1 + t2 + (u (x1 - x2))/c^2))/Sqrt[(c - u) (c + u)]
>
> In[16]:= s - dT
>
> Out[16]= (t1 - (u x1)/c^2)/Sqrt[1 - u^2/c^2] - (t2 - (u x2)/c^2)/Sqrt[
>  1 - u^2/c^2] + (-t1 + t2 - (u (-x1 + x2))/c^2)/Sqrt[1 - u^2/c^2]
>
> In[17]:= FullSimplify[s - dT]
>
> Out[17]= 0

[toc] | [prev] | [standalone]


Back to top | Article view | comp.soft-sys.math.mathematica


csiph-web