On Fri, 25 Apr 2008, Harald Geyer wrote:
> As you seem to have trouble getting a minimal example, my second guess
> is, that you just run out of memory in one way or the other. (Sorry
> don't know what clisp does in this case). Perhaps you get bitten by
> the oom killer?
I didn't know about that... sounds quite plausible, particularly as my
process was niced.
Another clue... by changing my notation in a way that simplified the
equations a bit, I managed to coax solve to give an error message when
it failed:
Maxima encountered a Lisp error:
string too long: desired length 6553600 exceeds the supported maximum
length
So... looks like the length of the algebraic expression I'm trying to
solve (which was generated by a sequence of previous solves and
substs) is just too much for clisp.
I can get a shorter, soluble expression out by using zeroth, instead
of first, order Taylor expansions in some (hopefully) small
parameters.
--
Thanks very much,
Dan