Re: Don't understand behaviour of Solve[]
- To: mathgroup at smc.vnet.net
- Subject: [mg58570] Re: [mg58554] Don't understand behaviour of Solve[]
- From: Andrzej Kozlowski <akozlowski at gmail.com>
- Date: Fri, 8 Jul 2005 00:46:01 -0400 (EDT)
- References: <200507070935.FAA29377@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
From the documentation:
Solve gives generic solutions only. It discards solutions that are
valid only when the parameters satisfy special conditions. Reduce
gives the complete set of solutions.
(The meaning of generic is left a bit vague but basically solutions
are called generic if they are valid for infinitely many values of
the parameters and not valid for at most finitely many.)
So, for example:
Solve[{x + y == 3, x - y == 1}, {x, y}]
{{x -> 2, y -> 1}}}
while
Solve[{x + y == 3, x - y == 1}, x]
{}
The second set of equations has a solution for only when the
parameter y is 1, so the set of generic solutions for x is {}.
Note that
Reduce[{x + y == 3, x - y == 1}, x]
Out[7]=
y == 1 && x == 2
Here is another example illustrating the same thing:
Solve[Exp[x]*a == 2*Exp[x]*a, x]
{}
but
Reduce[Exp[x]*a == 2*Exp[x]*a, x]
a == 0
On the other hand:
Solve[x/a == 2*x/a, x]
{{x->0}}
This time the answer is valid only when a !=0
Reduce[x/a == 2*(x/a), x]
x == 0 && a != 0
but Solve ignores this because it only gives a "generic" answer.
Andrzej Kozlowski
Chiba, Japan
On 7 Jul 2005, at 18:35, Oliver Friedrich wrote:
> Hallo,
>
> I have a problem understanding the general behaviour of the Solve[]
> function.
>
> I have a set of equations with the variables a,b,c,d and k.
>
> Evaluating Solve[set,k] or Solve[set,{a,b}] returns with {},
> whereas Solve
> [set,{k,b,c,d}] returns solutions for k,b,c and d.
>
> My question: Why does Solve returns no solution when searching only
> for k
> for example and why is there a solution for several variables? I
> thought
> that giving a list of variables just means that I want to search
> for all of
> them. But the procedure doesn't seem to be independant from one
> variable to
> another.
>
> What kind of information contains my list of variables except my
> wish to
> solve for these?
>
> Many thanks
>
> Oliver friedrich
>
>
- References:
- Don't understand behaviour of Solve[]
- From: Oliver Friedrich <oliver.friedrich@tzm.de>
- Don't understand behaviour of Solve[]