Re: Why does Solve give me no solutions for this in Version 8.0.1?
- To: mathgroup at smc.vnet.net
- Subject: [mg119968] Re: Why does Solve give me no solutions for this in Version 8.0.1?
- From: Peter Breitfeld <phbrf at t-online.de>
- Date: Sat, 2 Jul 2011 05:01:48 -0400 (EDT)
- References: <iuj54s$nnr$1@smc.vnet.net>
Phil J Taylor wrote:
> This system of equations for the Magic Hexagon is indeterminate.
> Solve still provides useful information in Version 6.0
>
> ClearAll[a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s];
> eqns = {
> a + b + c + d + e + f + g + h + i + j + k + l + m + n + o + p + q + r + s
> - 190 == 0,
> a + b + c - 38 == 0,
> a + d + h - 38 == 0,
> a + e + j + o + s - 38 == 0,
> b + e + i + m - 38 == 0,
> b + f + k + p - 38 == 0,
> c + f + j + n + q - 38 == 0,
> c + g + l - 38 == 0,
> d + e + f + g - 38 == 0,
> d + i + n + r - 38 == 0,
> g + k + o + r - 38 == 0,
> h + i + j + k + l - 38 == 0,
> h + m + q - 38 == 0,
> l + p + s - 38 == 0,
> m + n + o + p - 38 == 0,
> q + r + s - 38 == 0
> };
>
> Join[
> Solve[eqns, b], Solve[eqns, d], Solve[eqns, g],
> Solve[eqns, m], Solve[eqns, p], Solve[eqns, j],
> Solve[eqns, r], Solve[eqns, e], Solve[eqns, f],
> Solve[eqns, i], Solve[eqns, k], Solve[eqns, n],
> Solve[eqns, o], Solve[eqns, a], Solve[eqns, c],
> Solve[eqns, h], Solve[eqns, l], Solve[eqns, q],
> Solve[eqns, s]]
>
> Out[1]: {{b -> j + n + o}, {d -> j + k + o}, {g -> i + j + n},
> {m -> f + j + k}, {p -> e + i + j}, {j -> -38 + d + g + r},
> {r -> -38 + h + l + m + p}, {e -> -38 + h + k + q + r},
> {f -> -38 + i + l + r + s}, {i -> -38 + f + p + q + s},
> {k -> -38 + e + m + q + s}, {n -> -38 + g + h + k + l},
> {o -> -38 + d + h + i + l}, {a -> -38 + i + m + n + q + r},
> {c -> -38 + k + o + p + r + s}, {h -> -38 + n + o + p + r + s},
> {l -> -38 + m + n + o + q + r}, {q -> -38 + g + k + l + o + p},
> {s -> -38 + d + h + i + m + n}}
>
> Version 8.0.1 returns {}.
You can get back the behaviour of Solve (v.6), setting the Option
Method, which in 8.0.1 defaults to Automatic:
SetOptions[Solve,Method->"Legacy"]
Now Solve will give you the expected results.
--
_________________________________________________________________
Peter Breitfeld, Bad Saulgau, Germany -- http://www.pBreitfeld.de