I would like to see some feature in that direction, since all the *.MW new interface does not properly run on my old Win ME - and if it at least opens the sheets, it needs endless time for displaying it. Thus I shoot it to print as *.ps files - which gives tremendous file size and converting that to pdf neither helps much nor does it give 'readable' output (it looks terrific). What I say: the workaround Maple -> postscript -> pdf does not neccessary give reasonable results at all.

one would use a numerical root finder from common libraries (i do not have them at hand and may be they do not find all roots directly. You may look for Netlib C and Brent (1 dim root finder) or dig for it in the GSL library Or code some explicitely in Maple and compile it (which will give you much for testing)

one would use a numerical root finder from common libraries (i do not have them at hand and may be they do not find all roots directly. You may look for Netlib C and Brent (1 dim root finder) or dig for it in the GSL library Or code some explicitely in Maple and compile it (which will give you much for testing)

It is almost impossible to answer in detail, but if you use Maple specific constructs you will have to eliminate them first: for globals there are globals in C (but take care ...) and a seq or list usually can be handled as array or pointer. If you want to use compile check the help pages before, it is heavily restricted concerning arguments (do not have it here, but esentially you can have only quite simple constructions and no functions as arguments). In short: you are supported to bring things to C, but it is not automatically.

i am missing a 'built in' method as well, but i would guess it is quite 'expensive' to test the branching behaviour numerically as general recipe.

i am missing a 'built in' method as well, but i would guess it is quite 'expensive' to test the branching behaviour numerically as general recipe.

you are right, sorry, i got it through regression/common options ... thx

you are right, sorry, i got it through regression/common options ... thx

Alec, Thx! Sometimes I want some key to press to switch to 'wizard mode' :-) Would be nice if the boyz and girlz from Maple update their docu ... Edited to add: what I mean is that the Optimization is really a very nice thing to have within Maple. But the documentation is too short and very hard to follow (at least for me). It would be quite helpfull to have it brushed up - especially for using the so-called maxtrix versions (if the objective function is purely numerical, say it is given through a compiled external library). At least some more examples might help (including the various options and constraints).

Alec, Thx! Sometimes I want some key to press to switch to 'wizard mode' :-) Would be nice if the boyz and girlz from Maple update their docu ... Edited to add: what I mean is that the Optimization is really a very nice thing to have within Maple. But the documentation is too short and very hard to follow (at least for me). It would be quite helpfull to have it brushed up - especially for using the so-called maxtrix versions (if the objective function is purely numerical, say it is given through a compiled external library). At least some more examples might help (including the various options and constraints).

sorry for being not clear enough: i save the sheet with assumtions not shown, close it and with opening it again the tildes appear. interface(showassumed=0) is in the ini file so should act in any case.

sorry for being not clear enough: i save the sheet with assumtions not shown, close it and with opening it again the tildes appear. interface(showassumed=0) is in the ini file so should act in any case.

So this is a kind of parametric solution ... I have had a look at the link for LOCA and it may be you can compile it to call it as runtime library (perhaps also one of the others). If it is already existent as runtime lib, then may be you only need to write a wrapper to work with it (if it has not the correct calling conventions).

out of curiosity: what is that?

Not quite clear to me what you mean: do have several, discrete x and want to plot the pairs x,f(x) or do you have several pairs of x,y and want to plot y,f(x) (to compare y given and f(x) computed)? With pointplot you can plot pairs of values, cf the online help. So a way is to produce a list of pairs and then use pointplot. Depending on your matrix (a 2-dim array?) you do something like that

xyList:= [seq([M[i,1], M[i,2]],i=1..n)]:
  # you have n rows
  #  M[i,1], M[i,2]  are the entries x,y in row i,
  # [M[i,1], M[i,2]] is a 2-dim point [x,y]
  # seq gives the sequence of all that points
  # and enclosing that in brackets gives the input for pointplot:

plots[pointplot](xyList);

In the second case use

yfxList:= [seq([M[i,2], f(M[i,1])],i=1..n)]