Ah now I see that StartServer is only used for hpc mode. One of those bits of documentation observed in retrospect.

@Carl Love 

StartServer(numnodes = 1);
Error, (in Grid:-Server:-StartServer) no toolbox matching Grid found

Though interestingly Setup() allows me "local", Status() works and Launch() of the toy example appears to work. Perhaps somehow the server is already running. Time to test further.

@acer Ah, it appears that is extra cost.

As I do not fall within "Student" or "Academic" or "Government" it appears I would be "Commercial", so the purchase price would be about $CAD 240. That is unfortunately beyond my budget as a volunteer.

Could you give some sample a, b, p for us to test with?

@Carl Love You can typeset in 2D with

T := 1/`Δx` * Sum( b[i,n] * (f[n]-f[i]), n='{0,1,2,3,4,5,6}\{i}');

with the usual warnings about how you should be using add() instead of Sum() for addition of definite terms.

This must be used carefully, such as

eval(eval(T,1),i=2)

or else the set minus will end up operating on the symbolic i rather than on the i to be substituted.

If you try to use n in '{$0..6}\{i}' then the {$0..6} will not be evaluated early enough and the address of the `$`(0..6) structure will be compared to the address of the integer 2 (the i after substitution). But you can at least code it more smoothly (sort of) even if the display comes out like the above:

T := 1/`Δx` * Sum( b[i,n] * (f[n]-f[i]), n ='`minus`'(${0..6},{i}));

I think if I were doing it more often I might use something like

T := 1/`Δx` * Sum( b[i,n] * (f[n]-f[i]), n='{Rn}\{i}');

eval(eval(T,1),[Rn=$0..6,i=2])

but then I would be tempted to optimize to

T := 1/`Δx` * Sum( b[i,n] * (f[n]-f[i]), n='`0..6≠i`');

eval(eval(T,1),`0..6≠i`={$0..6)\{i})

@acer You are right, it is a list of list being returned by solve()

op(map(rhs~, solve(...))) will work, at least on newer Maple. Or to get fancier,

(op@rhs~~@solve)({x^2+y=1,x-y=5},[x,y],explicit))

Though I do admit that an expression sequence of lists is seldom to my taste.

More verbosely, and usable with older Maple,

op(map(v->map(rhs,v),solve({x^2+y=1,x-y=5},[x,y],explicit)))

Is "als Zeigergröße" something about vectors? Or something about magnitude possibly?

The source appears to be this German .pdf
Automated translation via Google at this URL

In your forms, you take Re(s) and Re() of some expressions that are provably real if s is real, or Re() of some expressions that are provably real between s = 0 and the pole. The expressions can, in some places, be simplified if assumptions can be placed on them. It would help if you could identify the assumptions at each stage, or at least the assumptions at the stage you are looking for speedup on. For example the M and g you time on are both positive: may we add to the assumptions that M and g are real and non-negative ?

Better:

L := [9, 2, -3, 4, -3, 4, 4];

with(ListTools):
T0 := Classify(p->L[p], [$1..nops(L)]);
ic := map(min@rhs, [indices(T0,indexorder,pairs)]);


I am still working on ia

@Kitonum MATLAB's unique() command can have multiple outputs, with the extra ones being indices. It is not clear from the question whether MATLAB's "ia" or "ic" is desired (or both).

Are the elements numeric? Extended numeric? symbols? expressions that will be "obviously" the same or not? Do trig identities need to be taken into account, so that sin(x)^2 and 1 - cos(x)^2 should be considered to be the same?

@Rouben Rostamian  I see reports that in more complicated plots, especially those that include added plots or images or transparency, that jpeg has been the only functional output format in each version released since roughly Maple 14. I do not happen to recall if the people tested png, but .eps is the one described as being the worst affected, with .ps more functional but still important deficiencies. The details of the problems have changed a bit with the releases. Improvements are promised for some of the forthcoming releases.

See here for some of the problems described by erik10

Remember that in Maple, RandomVariable() creates a data structure describing the distribution, but does not sample the distribution. The output of RandomVariable() looks like (for example) _R103 . You need to apply Sample() to the random variable in order to get particular random values.

The Maplesoft provided code for FormatMessage hot-patches itself to become a call into an external library, so I won't be able to extract anything useful from it. Ah well.