You could reflect a plot with respect to the x axis like this: p:=plot(cos): p; q:=plottools[reflect](p,[[-1,0],[1,0]]): q; Is this what you need?

You cannot use assign(solve1) here, because it does not generate a function. (Try evaluating dp[n] afterwards for other arguments than l and x!). Instead, use dp[n]:=unapply(rhs(solve1),l,x) I didn't check the rest of the solution, though.

Could you provide the full code that you entered, so that we can try to reproduce the error message? Thanks!

I think this is a known problem. Please see here.

From Maple 9.5 on, the int command recognizes such integrands, so you could use Diff(int(%,x),x), but there should be some more elegant approach...

This sounds like a bug to me, because it worked in earlier versions. A workaround for Maple 10 is convert(convert(Diff(f(x)+g(x),x)-Diff(g(x),x),D),Diff)

You said you are running Maple 8 and Matlab 6.5. That is most probably the source of the problem. Please note that the link in Maple 8 supports Matlab 6 only. This is documented at the top of the ?Matlab,setup help page. Maple 9 and 9.5 support Matlab 6.5, and Maple 10 supports Matlab 7 and 6.5, depending on the platform. Unsupported combinations may or may not work. For example, Maple 8 works with Matlab 5.3, but with 6.5, we have seen problem reports from other customers.

I don't know about your first question, but the second is easy: just use plotsetup(window);. To switch back to the default, use plotsetup(inline);. The same is available through the Display tab in the Tools/Options menu.

If you briefly describe what genvecs does, chances are good that you get replies from non-Mathcad-users like me.

Yes, such equations can be solved like this: rsolve({Y(t+1) + Y(t) = (-1)^t, Y(0)=2}, Y(t));. You might want to apply expand(%); to the result.

I think the easiest way is to load the Natural Units Environment. Please see ?Units,Natural and ?examples,NaturalUnits. The example worksheet also points out that explicit conversions may be preferrable if efficiency is an issue. BTW: I guess your input lines were meant to be x:=4*m and y:=5*m rather than x:=4(m) and y:=5(m), resp.. The latter variants are interpreted as constant functions, and automatically simplified to that constant value. In the Natural Units Environment, the output of the former variant will be displayed as x:=4[m] and y:=5[m], using square brackets.

Simply apply "Unix Network" update, called Maple1002UnixUpgrade.tar.gz. I have not applied this to my 64-bit Linux machine yet, but it worked that way the last time (10.00->10.01), and the installer script for 10.02 uses the same technique.

I'm not an expert on this topic, but I'd say that searching for such beasts on a single PC requires A LOT of patience, no matter what software you have. The largest known primes are usually those of Mersenne type, and according to the GIMPS Project, Maple 10's built-in list of Mersenne numbers is up-to-date. See ?numtheory[mersenne] for more information. In particular, numtheory[mersenne]([42]) will return the currently largest known prime. But even displaying it will require some time...

As a workaround, you could change the value for the 'lower' option to the Slider element in the exported .maplet file, using your favourite text editor. But make sure you keep a non-modified copy, because the edited file may not be readable from the Maplet Builder. This is a known limitation. If you create a Maplet the old-fashioned way (remember, the Maplet Builder is relatively new), you can use negative numbers from the very beginning, of course.

Robert Israel has a solution in his Maple Advisor Database. Please see labelledcontourplot.