Thanks again for reporting this problem.
I expect we'll fix this in solve, but in the meantime I found that simply
omitting the variable list gives the correct results, that is:
s1b:solve([e2,e3]);
Sorry for the inconvenience.
-s
On Tue, Jul 20, 2010 at 16:27, Adam Majewski <adammaj1 at o2.pl> wrote:
> Hi,
>
> It is fragment of code, where I try to solve system
> of 2 equations to find x1b and y1b
>
>
> ====================begin ================
> /* outer circle with center at origin */
> r0:400; /* radius */
> x0:0; /* realpart of center */
> y0:0; /* imagpart of center */
>
> /* I want to pack in 3 equall circles :
> I can compute its radius : */
>
> a1:(1 + 2 / sqrt(3)); /* http://www2.stetson.edu/~efriedma/cirincir/ */ /
> * float is needed for solve */
> r1:r0/a1;
> x1a:0;
> y1a:(r0-r1);
>
> e2:(x0-x1b)^2+(y0-y1b)^2 -(r0-r1)^2=0; /* 0 and 1a */
> e3:(x1a-x1b)^2+(y1a-y1b)^2 -(r1+r1)^2=0; /* 1a and 1b */
> s1b:solve([e2,e3],[x1b,y1b]);
> =============== end =========================
>
>
> s1b is empty = no results.
>
> When I change
> a1:(1 + 2 / sqrt(3));
> to
> a1:float(1 + 2 / sqrt(3));
>
> It gives good solutions.
>
> Adam
>
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima
>