Choose view that doesn't include vertical lines with negative x, something like view=[0..6,0..1.5]. I don't think that I chose framed axes and 500 numpoints - I just modified that guy's example. They can be added to the options in the display command. Many other options can be added, see ?plot,option __________ Alec Mihailovs http://mihailovs.com/Alec/

Choose view that doesn't include vertical lines with negative x, something like view=[0..6,0..1.5]. I don't think that I chose framed axes and 500 numpoints - I just modified that guy's example. They can be added to the options in the display command. Many other options can be added, see ?plot,option __________ Alec Mihailovs http://mihailovs.com/Alec/

For example,

f := 2*a*tan(t):

coeff(f,tan(t));

                                 2 a

remove(type,f,function);

                                 2 a

subsindets(f,function,1);

                                 2 a

eval(f,tan=1);

                                 2 a

select(type,f,atomic);
                                 2 a

_________ Alec Mihailovs http://mihailovs.com/Alec/

For example,

f := 2*a*tan(t):

coeff(f,tan(t));

                                 2 a

remove(type,f,function);

                                 2 a

subsindets(f,function,1);

                                 2 a

eval(f,tan=1);

                                 2 a

select(type,f,atomic);
                                 2 a

_________ Alec Mihailovs http://mihailovs.com/Alec/

For example,

f:=2*a+5*b+3.4*c:
 
`+`(coeffs(f));
                                 10.4

or

subsindets(f,name,1);
                                 10.4

or

map2(op,1,f);
                                 10.4

or

map(lcoeff,f);
                                 10.4

or

norm(f,1);
                                 10.4

(this one is shorter, but it is inefficient and gives the sum of coefficients only if the coefficients are non-negative.) __________ Alec Mihailovs http://mihailovs.com/Alec/

For example,

f:=2*a+5*b+3.4*c:
 
`+`(coeffs(f));
                                 10.4

or

subsindets(f,name,1);
                                 10.4

or

map2(op,1,f);
                                 10.4

or

map(lcoeff,f);
                                 10.4

or

norm(f,1);
                                 10.4

(this one is shorter, but it is inefficient and gives the sum of coefficients only if the coefficients are non-negative.) __________ Alec Mihailovs http://mihailovs.com/Alec/

Jan, Even if rotation on 90, 180, and 270 degrees can be done quite easily (as in that example), it is certainly good to have a command for that. What values for the second argument will be possible? Would that be possible to rotate on 30, or 60 degrees? What about reflections? And homothety? Crop? I am anxious to get Maple 11 to see what other editing operations are available. Wow, I just noticed View instead of Preview. Does that mean that the picture (or plot) will be displayed in all available space (and without axes)? Thank you, __________ Alec Mihailovs http://mihailovs.com/Alec/

That would give wrong values to the coefficients (the order is wrong),

a,b,c := coeffs(rhs(eq_fit),x);

                         a, b, c := -4, 3, 1

The following works in this example,

c,b,a:=coeffs(rhs(eq_fit),x);

                         c, b, a := -4, 3, 1

but one can't depend on that, the coefficients in coeffs are not ordered according to the degrees (of x in this case). The following is more robust,

a,b,c:=seq(coeff(rhs(eq_fit),x,i),i=2..0,-1);

                         a, b, c := 1, 3, -4

Patmatch also can be used,

patmatch(rhs(eq_fit),a::integer*x^2+b::integer*x+c::integer,S);
                                 
                               true

S;

                        [a = 1, b = 3, c = -4]

but I don't think that that is his (or her) problem. It looks as if he (or she) is trying to find a way to calculate these coefficients manually, step by step, without using fit commands. __________ Alec Mihailovs http://mihailovs.com/Alec/

That would give wrong values to the coefficients (the order is wrong),

a,b,c := coeffs(rhs(eq_fit),x);

                         a, b, c := -4, 3, 1

The following works in this example,

c,b,a:=coeffs(rhs(eq_fit),x);

                         c, b, a := -4, 3, 1

but one can't depend on that, the coefficients in coeffs are not ordered according to the degrees (of x in this case). The following is more robust,

a,b,c:=seq(coeff(rhs(eq_fit),x,i),i=2..0,-1);

                         a, b, c := 1, 3, -4

Patmatch also can be used,

patmatch(rhs(eq_fit),a::integer*x^2+b::integer*x+c::integer,S);
                                 
                               true

S;

                        [a = 1, b = 3, c = -4]

but I don't think that that is his (or her) problem. It looks as if he (or she) is trying to find a way to calculate these coefficients manually, step by step, without using fit commands. __________ Alec Mihailovs http://mihailovs.com/Alec/

For example,

X,Y:=<1,2,3,4>,<0,6,14,24>:
Statistics:-NonlinearFit(a*t^2+b*t+c,X,Y,t,output=parametervalues);

  [a = 0.99999999999999988, b = 3.00000000000000088,

        c = -4.00000000000000088]

convert(%,rational);

                        [a = 1, b = 3, c = -4]

_________ Alec Mihailovs http://mihailovs.com/Alec/

For example,

X,Y:=<1,2,3,4>,<0,6,14,24>:
Statistics:-NonlinearFit(a*t^2+b*t+c,X,Y,t,output=parametervalues);

  [a = 0.99999999999999988, b = 3.00000000000000088,

        c = -4.00000000000000088]

convert(%,rational);

                        [a = 1, b = 3, c = -4]

_________ Alec Mihailovs http://mihailovs.com/Alec/

Just op it. __________ Alec Mihailovs http://mihailovs.com/Alec/

Just op it. __________ Alec Mihailovs http://mihailovs.com/Alec/

draw3d is the best choice in many cases for drawing graphs. And the plot can be rotated using a mouse. __________ Alec Mihailovs http://mihailovs.com/Alec/

draw3d is the best choice in many cases for drawing graphs. And the plot can be rotated using a mouse. __________ Alec Mihailovs http://mihailovs.com/Alec/