Thank you for your help. It works perfect.

Thanks for the example, but perhaps I have not discribed my problem in the right way. I know how to display the plots in a row or column with display.
But my problem is, that I have a plot and a textplot and I want to have them in the same picture. So I want to have
a row of plots/pictures , but in the first location of the row I want to have the picture of the plot and the textplot together.

If I try it this way I do not get an error, but the textplot is missing.

restart:
with(plots):
p1:=plot(x^2):
p2:=textplot([0,0, "Minimum"]):
p3:=plot(x-1):
L:=Array(1..1,1..3):
L[1,1]:=p1, p2;
L[1,2]:=p3:
L[1,3]:=p3:
display(L);

Thank you for your help.
Helena

Hi Arcer,

thank you a lot for the help and the links. I think I know now how to do it.
 

Helena

Thank you for your help Acer.

But if NLP-Solve works like that I have another question.
In the case above I see that ist doesn't make a problem, that the solution for y[2] is near to zero, but negative,
because the function F2 is continuous in x[2]=6.33, y[2]=0.
My question is: How does NLP-Solve handle such tolerances if it is not continuous at that point. If the function
would be discontinuous in x[2]=6.33, y[2]=0 it could perhaps be a a really big difference whether it is a point near to zero, but negative or a point near to zero but positive.

Thank you for your help
Helena

This was exactly what I was searching for.

Thanks for your help

I want to prove whether the set is empty, because I want to avoid an Error Message in the NLPSolve command. So I can't use this way to prove whether the set is empty. I need a way that doesn't give me an Error Massage, because my procedure stops when I get such a massage.
Or is there a way to avoid that the procedure stops when I get an Error???

Thanks for your Help

Helena

The routine rtable_scanblock works perfect.

Thank you for your help Arcer and Scott