Skip to main content

  | मौसम…

411 समाचार
Public archive edition · Attributed source reporting with original dates · AI drafts are excluded

411 REFERENCE · Wikibooks

Fractals/Multibrot sets

Reference material by Wikibooks contributors. This retains its source identity and is not 411 news reporting.

Julia sets for polynomial functions " The dynamics of polynomials is much better understood than the dynamics of general rational maps" due to the Bottcher’s theoremON THE NOTIONS OF MATING by CARSTEN LUNDE PETERSEN AND DANIEL MEYER z(1 + z)^d Map has d petals. "Although the parabolic point at z = 0 has only one petal, the map f also has a preparabolic critical point b = −1 of local degree d. Thus f has d petals at b. " dimension and conformal dynamics II: Geometrically finite rational maps by Curtis T McMullen For example "the filled Julia set of f (z) = z(1 + z) 3 has three petals at z = −1." ( Curtis T McMullen ) mz(1+z^(k*n)) "The condition describes a set of lines through the origin. Note that the map g leaves these lines invariant. Then is the composition of g with the periodic rotation "COMPLEX ANALYTIC DYNAMICS ON THE RIEMANN SPHERE BY PAUL BLANCHARD z(1-z^q) z^n+ z + c Example f(z) = z^3 + z + 0.6*IInverse Iteration Algorithms for Julia Sets by Mark McClure f(z) = z^6+A*z +c ; degree 6, A=(-34603008+0*i)*2^-25, c=(-262144+0i)*2^-25 from another parameter space slice showing 5 attracting period-2 cycles by marcm200fractalforums.org: julia-sets-true-shape-and-escape-time London towers" Period-3 cycle and Period-18 cycle , z^5+A*z+c c=(638976+7536640*i) * 2^-25 A=(37094161+719769*i) * 2^-25 by marcm200 fractalforums.org: julia-sets-true-shape-and-escape-time z^n + c^p Z = Zn + cp L = (m - 1) * p z^n + c = multibrot "the mulitbrot family, ... they all have a single critical point, namely the origin" Mark McClure Mark McClure - comment Multibrot sets wikipedia : Multibrot_set High-Order Mandelbrot and Julia Sets Written by Christopher Thomas. Images and Perl programStephen Haas : The Hausdorff Dimension of the Julia Set of Polynomials of the Form z^d+cvideo on youtube by rrwick L = m - 1 Compare with it's inverse parameter plane : Z^n + 1/c Desenvolupament_de_fractals_mitjan How to compute powers of z : Fortan source code : ! Fortran program by P.M.J. Trevelyan ! http://philiptrevelyan.co.uk/ PROGRAM FRACTAL IMPLICIT NONE INTEGER I,J,ITERATION,N,M PARAMETER(N=2000,M=50) REAL*8 U,V,X,Y,P,Q OPEN(99,FILE='Fractal_quad.dat') 25 FORMAT(2F9.5,I3) DO 10, I=1,N DO 20, J=1,N C Define first point z(n)=U+iV and k=X+iY U=0.D0 V=0.D0 X=I*3.2D0/(N-1.D0)-2.1D0 Y=J*2.8D0/(N-1.D0)-1.4D0 DO 30, ITERATION=1,M C Calculate z(n+1) = z(n)**2 + k where z(n+1)=P+iQ P=U**2-V**2+X Q=2.D0*U*V+Y U=P V=Q C If |z|>2 stop iterating If (U**2+V**2.GT.4.D0) GOTO 100 30 CONTINUE 100 WRITE(99,25) X,Y,ITERATION 20 CONTINUE 10 CONTINUE STOP END Parameter planes for positive integer powers : Negative powers When d is negative the set surrounds but does not include the origin. There is interesting complex behaviour in the contours between the set and the origin, in a star-shaped area with (1 − d)-fold rotational symmetry. The sets appear to have a circular perimeter, however this is just an artifact of the fixed maximum radius allowed by the Escape Time algorithm, and is not a limit of the sets that actually extend in all directions to infinity. z^2+c See complex quadratic polynomials z^3 + c It can be computed by :De la génération des fractales Théorie et Applications John Bonobo Examples: c= -0.040000000000000036 + I * -0.78wolfram : julia-set-explorations z^5 + c z^10 + c zn + 1/c It is an inverted parameter plain of z^n + c. Number of vertices : V = (n - 1) z^n + m*z^(-d) McMullen maps : where : n and d are >=1 "These maps are known as `McMullen maps', since McMullen C. McMullen, Automorphisms of rational maps. In `Holomorphic Functions and Moduli I', 31-60, Springer-Verlag, 1988. first studied these maps and pointed out that when (n: d) = (2: 3) and m is small, the Julia set is a Cantor set of circles." Hyperbolic components of McMullen maps Weiyuan Qiu, Pascale Roesch, Xiaoguang Wang, Yongcheng Yin z^n + m*z Description "Visualizing the complex dynamics of families of polynomials with symmetric critical points" by Ning Chena, Jing Suna, Yan-ling Suna, Ming Tangb. Chaos, Solitons & Fractals. Volume 42, Issue 3, 15 November 2009, Pages 1611–1622A topological characterisation of holomorphic parabolic germs in the plane Fr ́ed ́eric Le Roux z^2+m*z See complex quadratic polynomial z^3+m*z dynamic planes z^3 + z It can be found using Maxima CAS : (%i2) z:zx+zy*%i; (%o2) %i*zy+zx (%i6) realpart(z+z^3); (%o6) -3*zx*zy^2+zx^3+zx (%i7) imagpart(z+z^3); (%o7) -zy^3+3*zx^2*zy+zy Finding roots and its multiplicity : so root z=0 has multiplicity 3. (%i1) z1:z^3+z; (%o1) z^3+z (%i2) solve(z1=z); (%o2) [z=0] (%i3) multiplicities; (%o3) [3] It means that there is a flower with 2 petals around fixed point z=0. Compare figures : by Alessandro RosaOne approach to the digital visualization of hedgehogs in holomorphic dynamics Alessandro Rosa, fig 5.13 page 26 by Xavier Buff and Adam L. EpsteinA parabolic Pommerenke-Levin-Yoccoz inequality by Xavier Buff and Adam L. Epstein, fig 1 page 4 z^4 + mz How to compute iteration : (%i1) z:zx+zy*%i; (%o1) %i*zy+zx (%i2) m:mx+my*%i; (%o2) %i*my+mx (%i3) z1:z^4+m*z; (%o3) (%i*zy+zx)^4+(%i*my+mx)*(%i*zy+zx) (%i4) realpart(z1); (%o4) zy^4-6*zx^2*zy^2-my*zy+zx^4+mx*zx (%i5) imagpart(z1); (%o5) -4*zx*zy^3+4*zx^3*zy+mx*zy+my*zx See also f(z) = c(z^4-4z). It is a family 4.1 in program Mandel by Wolf JungMandel: software for real and complex dynamics by Wolf Jung ( see main menu / New / 4. Quartic polynomials / 4.1 ) Here c = -m/4 and Mandelbrot set is rotated by 180 degrees. Parameter plane thumbnail|right|Parameter plane and Mandelbrot set for f(z) = z^4 + m*z Period 1 components Maxima CAS code : (%i1) f:z^4+m*z; (%o1) z^4+m*z (%i2) e1:f=z; (%o2) z^4+m*z=z (%i3) d:diff(f,z,1); (%o3) 4*z^3+m (%i4) e2:d=w; (%o4) 4*z^3+m=w (%i5) s:eliminate ([e1,e2], [z]); (%o5) [-(m-w)*(w+3*m-4)^3] (%i6) s:solve([s[1]], [m]); (%o6) [m=-(w-4)/3,m=w] It means that there are 2 period 1 components : one with radius = 1 and center =0 ( m=w ) second with radius 1/3 and center=4/3 ( m=-(w-4)/3 ) Dynamic planes z^4+z thumbnail|right|Julia set for f(z) = z^4 + z Finding roots and its multiplicity : so root z=0 has multiplicity 4. (%i1) z1:z^4+z; (%o1) z^4+z (%i2) solve(z1=z); (%o2) [z=0] (%i3) multiplicities; (%o3) [4] It means that there are 3 petals around fixed point z=0 Complex dynamics, Lennart Carleson, Theodore W. Gamelin, Springer, 1993, 978-0-387-97942-7. Page 40, Figure 2. How to compute iteration : (%i17) z:x+y*%i; (%o17) %i*y+x (%i18) realpart(z+z^4); (%o18) y^4-6*x^2*y^2+x^4+x (%i19) imagpart(z+z^4); (%o19) -4*x*y^3+4*x^3*y+y z^4-iz right|thumb|Critical orbits for f(z)=z^4-iz First compute multiplier for internal angle=3/4 : (%i1) m:exp(2*%pi*%i*3/4); (%o1) -%i Then find how to compute iteration : (%i1) z:x+y*%i; (%o1) %i*y+x (%i2) z1:z^4-%i*z; (%o2) (%i*y+x)^4-%i*(%i*y+x) (%i3) realpart(z1); (%o3) y^4-6*x^2*y^2+y+x^4 (%i4) imagpart(z1); (%o4) -4*x*y^3+4*x^3*y-x It is a parabolic Julia set with 12 petal flower F. Bracci, Local holomorphic dynamics of diffeomorphisms in dimension one. Contemporary Mathematics 525, (2010), 1-42. Notes of the PhD course given in A.A 2007/08. Critical points : (%i12) s:GiveListOfCriticalPoints(f(z)) (%o12) [0.31498026247372*%i-0.54556181798586,-0.62996052494744*%i,0.31498026247372*%i+0.54556181798586] (%i13) multiplicities (%o13) [1,1,1] (%i14) length(s) (%o14) 3 with arguments in turns : [0.41666666666667,0.75,0.083333333333334] = [5/12 , 9/12, 1/12] Attracting vectors Multiplier of fixed point −i is a fourth root of unity ( q=4), thus we examine 4-th iteration : (%i1) z1:z^4-%i*z; (%o1) z^4-%i*z (%i2) z2:z1^4-%i*z1; (%o2) (z^4-%i*z)^4-%i*(z^4-%i*z) (%i3) z3:z2^4-%i*z2; (%o3) ((z^4-%i*z)^4-%i*(z^4-%i*z))^4-%i*((z^4-%i*z)^4-%i*(z^4-%i*z)) (%i4) z4:z3^4-%i*z3; (%o4) (((z^4-%i*z)^4-%i*(z^4-%i*z))^4-%i*((z^4-%i*z)^4-%i*(z^4-%i*z)))^4-%i*(((z^4-%i*z)^4-%i*(z^4-%i*z))^4-%i*((z^4-%i*z)^4-%i*(z^4-%i*z))) (%i6) taylor(z4,z,0,20); (%o6)/T/ z+(-76*%i-84)*z^13+(-36*%i+720)*z^16+(1812*%i-2556)*z^19+... Next term after z is a : (-76*%i-84)*z^13 so here : k=13 and n=m*q = k-1 = 12 a = -76*%i-84 Attracting vectors satisfy : so here : One can solve it in Maxima CAS : (%i14) s:map('float,s); (%o14) [1.007236559448514*%i+1.521106958434882,1.632845927320289*%i+0.81369898815363, 1.820935547602145*%i-0.11173896888541,1.521106958434882*%i-1.007236559448514, 0.81369898815363*%i-1.632845927320289, -0.11173896888541*%i-1.820935547602145,-1.007236559448514*%i-1.521106958434882, -1.632845927320289*%i-0.81369898815363,0.11173896888541-1.820935547602145*%i, 1.007236559448514-1.521106958434882*%i, 1.632845927320289-0.81369898815363*%i,0.11173896888541*%i+1.820935547602145] With arguments in turns : [0.093087406197659,0.17642073953099,0.25975407286433,0.34308740619766,0.42642073953099,0.50975407286433, 0.59308740619766,0.67642073953099,0.75975407286433,0.84308740619766,0.92642073953099,0.009754072864326] different then arguments of critical points. Thus critical orbits form distorted 12-arms star Find the fixed points : (%i1) f:z^4-%i*z; (%o1) z^4-%i*z (%i2) s:solve(f=z); (%o2) [z=((%i+1)^(1/3)*(sqrt(3)*%i-1))/2,z=-((%i+1)^(1/3)*(sqrt(3)*%i+1))/2,z=(%i+1)^(1/3),z=0] (%i4) multiplicities; (%o4) [1,1,1,1] (%i3) s:map(rhs,s); (%o3) [((%i+1)^(1/3)*(sqrt(3)*%i-1))/2,-((%i+1)^(1/3)*(sqrt(3)*%i+1))/2,(%i+1)^(1/3),0] (%i5) s:map('float,s); (%o5) [0.5*(%i+1.0)^(1/3)*(1.732050807568877*%i-1.0),-0.5*(%i+1.0)^(1/3)*(1.732050807568877*%i+1.0),(%i+1.0)^(1/3),0.0] (%i6) s:map(rectform,s); (%o6) [0.7937005259841*%i-0.7937005259841,-1.084215081491351*%i-0.29051455550725,0.29051455550725*%i+1.084215081491351,0.0] Compute the multiplier of fixed points : (%i7) d:diff(f,z,1); (%o7) 4*z^3-%i Check the stability of fixed points : (%i9) for z in s do disp(abs(ev(d))); 4.999999999999998 5.0 4.999999999999999 1 (%o9) done Point z=0 is a parabolic point. z^4 -z thumbnail|right|Critical points, orbits for f(z) = z^4 - z thumbnail|right|Julia set for f(z) = z^4 - z thumbnail|right|Julia set for f(z) = z^4 - z with various number of star branches ( animation ) - repelling petals It is a special case of polynomial from family : Here so internal angle is : (%i2) m:exp(2*%pi*%i/2); (%o2) -1 Because : it is a parabolic Julia set. Point is between two period one components ( root point ). Periodic points Point z=0 is a root of multiplicity seven for equation : One can check it in Maxima CAS using numerical : (%i1) z1:z^4-z; (%o1) z^4-z (%i2) z2:z1^4-z1; (%o2) (z^4-z)^4-z^4+z (%i3) eq2:z2-z=0; (%o3) (z^4-z)^4-z^4=0 (%i4) allroots(eq2); (%o4) [z=0.0,z=0.0,z=0.0,z=0.0,z=0.0,z=0.0,z=0.0,z=1.259921049894873, z=0.7937005259841*%i-0.7937005259841,z=-0.7937005259841*%i-0.7937005259841, z=1.084215081491351*%i-0.29051455550725,z=-1.084215081491351*%i-0.29051455550725, z=0.29051455550725*%i+1.084215081491351, z=1.084215081491351-0.29051455550725*%i,z=1.091123635971722*%i-0.62996052494744, z=-1.091123635971722*%i-0.62996052494744] (%i5) expand(eq2); (%o5) z^16-4*z^13+6*z^10-4*z^7=0 (%i6) factor(eq2); (%o6) z^7*(z^3-2)*(z^6-2*z^3+2)=0 and symbolic methods : (%i1) z1:z^4-z; (%o1) z^4-z (%i2) solve(z1=z); (%o2) [z=(2^(1/3)*sqrt(3)*%i-2^(1/3))/2,z=-(2^(1/3)*sqrt(3)*%i+2^(1/3))/2,z=2^(1/3),z=0] (%i3) multiplicities; (%o3) [1,1,1,1] (%i4) z2:z1^4-z1; (%o4) (z^4-z)^4-z^4+z (%i5) solve(z2=z); (%o5) [z=(2^(1/3)*sqrt(3)*%i-2^(1/3))/2,z=-(2^(1/3)*sqrt(3)*%i+2^(1/3))/2,z=2^(1/3),z=((%i+1)^(1/3)*(sqrt(3)*%i-1))/2,z=-((%i+1)^(1/3)*(sqrt(3)*%i+1))/2,z=(%i+1)^(1/3),z=(sqrt(3)*(1-%i)^(1/3)*%i-(1-%i)^(1/3))/2,z=-(sqrt(3)*(1-%i)^(1/3)*%i+(1-%i)^(1/3))/2,z=(1-%i)^(1/3),z=0] (%i6) multiplicities; (%o6) [1,1,1,1,1,1,1,1,1,7] Number of petals = 6 Complex dynamics, Lennart Carleson, Theodore W. Gamelin, Springer, 1993, 978-0-387-97942-7. Page 41 Atracting vectors Denominator of internal angle is so one have to check second iteration of function : (%i5) z1:z^4-z; (%o5) z^4-z (%i6) z2:z1^4-z1; (%o6) (z^4-z)^4-z^4+z (%i8) expand(z2); (%o8) z^16-4*z^13+6*z^10-4*z^7+z Next term after z is a -4z^7. Then : k = 7 and n=m*q = k-1 = 6 a = -4 Attracting vectors satisfy : so here : One can solve it using Maxima CAS : (%i10) s:solve(z^6=1/24); (%o10) [z=(sqrt(3)*%i+1)/(2^(3/2)*3^(1/6)),z=(sqrt(3)*%i-1)/(2^(3/2)*3^(1/6)),z=-1/(sqrt(2)*3^(1/6)),z=-(sqrt(3)*%i+1)/(2^(3/2)*3^(1/6)),z=-(sqrt(3)*%i-1)/(2^(3/2)*3^(1/6)),z=1/(sqrt(2)*3^(1/6))] (%i11) s:map(rhs,s); (%o11) [(sqrt(3)*%i+1)/(2^(3/2)*3^(1/6)),(sqrt(3)*%i-1)/(2^(3/2)*3^(1/6)),-1/(sqrt(2)*3^(1/6)),-(sqrt(3)*%i+1)/(2^(3/2)*3^(1/6)),-(sqrt(3)*%i-1)/(2^(3/2)*3^(1/6)),1/(sqrt(2)*3^(1/6))] (%i12) s:map('float,s); (%o12) [0.29439796075012*(1.732050807568877*%i+1.0),0.29439796075012*(1.732050807568877*%i-1.0),-0.58879592150024,-0.29439796075012*(1.732050807568877*%i+1.0),-0.29439796075012*(1.732050807568877*%i-1.0),0.58879592150024] (%i13) s:map(rectform,s); (%o13) [0.50991222566388*%i+0.29439796075012,0.50991222566388*%i-0.29439796075012,-0.58879592150024,-0.50991222566388*%i-0.29439796075012,0.29439796075012-0.50991222566388*%i,0.58879592150024] (%i14) s:map(carg_t,s); (%o14) [0.5235987755983/%pi,1.047197551196598/%pi,1/2,1-1.047197551196598/%pi,1-0.5235987755983/%pi,0] (%i15) s:map('float,s); (%o15) [0.16666666666667,0.33333333333333,0.5,0.66666666666667,0.83333333333333,0.0] So critical points lie on attracting vectors. Thus critical orbits tend straight to the origin under the iterationMark McClure in stackexchange questions : what-is-the-shape-of-parabolic-critical-orbit How to compute : (%i2) z:x+y*%i; (%o2) %i*y+x (%i3) realpart(z^4-z); (%o3) y^4-6*x^2*y^2+x^4-x (%i4) imagpart(z^4-z); (%o4) -4*x*y^3+4*x^3*y-y Critical points : s:GiveListOfCriticalPoints(f(z)) (%o8) [0.54556181798586*%i-0.31498026247372,-0.54556181798586*%i-0.31498026247372,0.62996052494744] These points has arguments in turns : 1/3, 2/3, 0 z^5 + m*z dynamic planes z^5 + z thumbnail|right|Julia set z+z^5. Image and src code Finding roots and its multiplicity : so root z=0 has multiplicity 5. It means that there is a flower with 4 petals One approach to the digital visualization of hedgehogs in holomorphic dynamics Alessandro Rosa around fixed point z=0. It How to compute : (%i23) z:x+y*%i; (%o23) %i*y+x (%i24) realpart(z+z^5); (%o24) 5*x*y^4-10*x^3*y^2+x^5+x (%i25) imagpart(z+z^5); (%o25) y^5-10*x^2*y^3+5*x^4*y+y In c programs one must use temporary variable so it can be : tempx = 5*x*y*y*y*y-10*x*x*x*y*y + x*x*x*x*x + x ; // temporary variable y = y*y*y*y*y -10*x*x*y*y*y + 5*x*x*x*x*y + y ; x=tempx; It can be optimized "...an escape time algorithm would take forever to generate that type of image, since the dynamics are so slow there. If you want resolution of 1/100, it would take roughly 2*10^8 iterates to move the point z0=0.01 to z=2 by iterating f(z)=z+z^5." ( Mark McClure stackexchange questions : what-is-the-shape-of-parabolic-critical-orbit "This picture shows the Julia Set of f(z) = z + z^5 which has an indifferent fixed point at z = 0. ( f(0) = 0 and f ' (0) = 1 .) The 4 lines : Re z = 0 and Im z = 0 and Re z = Im z and Re z = -Im z are invariant under iteration of f. On Im z = 0: f(x) = x + x^5 On Re z = 0: f(ix) = ix + (ix)^5 = ix + i^5 x^5 = i(x+x^5) On Re z = Im z , f(z) = r e^(i pi/4) + r^5 e^(i 5 pi/4) = e^ (i pi/4)(r - r^5) On Re z = -Im z, f(z) = = r e^(i 3pi/4) + r^5 e^(i 15 pi/4) = e^ (i 3pi/4)(r - r^5) Using one dimensional analysis it is easily shown that f(x) = x + x^5 has a repelling fixed point at x = 0 and f(x) = x - x^5 has an attracting fixed point at x = 0. Thus along the four invariant lines 0 is attracting on the first two and repelling on the second two. The points repelled from 0 are shown in shades of blue, while those attracted to 0 are shown in shades of brown. 0 has four attracting petals, which are in shades of brown. (A simply connected region C is a petal for an indifferent fixed point p if p is contained in the boundary of C and for each z in C, F^n(z) -> p (see Devaney - 1987)" Fractals by Anne Burns, Department of Mathematics, C.W. Post Campus, Long Island University 6 degree general case z^6+A*z+c c=(-6145144-20171676*i) * 2^-25 = (-6145144 - 20171676 i)/2^25 = -0.1831395626068115234375 - 0.60116279125213623046875 i Critical points: using WolframAlfa List: z = -0.454407 + 0.0918858 I z = -0.227808 - 0.403772 I z = -0.0530308 + 0.460561 I z = 0.313614 - 0.341431 I z = 0.421632 + 0.192756 I Higher precision +0.4216319827875524 +0.1927564710317439*%i -0.4544068504035357 +0.09188580053693407*%i -0.2278080284907348 -0.4037723222177803*%i +0.3136137458861571 -0.3414308193839966*%i -0.05303084977943909 +0.4605608700330989*%i z^6+m*z dynamical plane z6+z on plane [-1.2;1.2]x[-1.2;1.2]. It has 5 petals Some Julia sets 2 by Michael Becker z^14 + m*z dynamic plane z^14 - z thumbnail|right|Julia set for f(z) = z^14 - z . Image and c src code thumbnail|right|Critical orbit for f(z) = z^14 - z . Image and Maxima CAS src code How to compute iteration : /* Maxima CAS session */ (%i1) z:x+y*%i; (%o1) %i*y+x (%i2) z1:z^14-z; (%o2) (%i*y+x)^14-%i*y-x (%i3) realpart(z1); (%o3) -y^14+91*x^2*y^12-1001*x^4*y^10+3003*x^6*y^8-3003*x^8*y^6+1001*x^10*y^4-91*x^12*y^2+x^14-x (%i4) imagpart(z1); (%o4) 14*x*y^13-364*x^3*y^11+2002*x^5*y^9-3432*x^7*y^7+2002*x^9*y^5-364*x^11*y^3+14*x^13*y-y f(z)=z^14-z, on [-1,2;1,2]x[-1,2;1,2] has 26 petals. Compare with image by Michael Becker. Fixpunkte und Periodische Punkte by Michael Becker How to find fixed points : (%i1) z1:z^14-z; (%o1) z^14-z (%i2) solve(z1=z); (%o2) [z=2^(1/13)*%e^((2*%i*%pi)/13),z=2^(1/13)*%e^((4*%i*%pi)/13), z=2^(1/13)*%e^((6*%i*%pi)/13),z=2^(1/13)*%e^((8*%i*%pi)/13), z=2^(1/13)*%e^((10*%i*%pi)/13),z=2^(1/13)*%e^((12*%i*%pi)/13), z=2^(1/13)*%e^(-(12*%i*%pi)/13),z=2^(1/13)*%e^(-(10*%i*%pi)/13), z=2^(1/13)*%e^(-(8*%i*%pi)/13),z=2^(1/13)*%e^(-(6*%i*%pi)/13), z=2^(1/13)*%e^(-(4*%i*%pi)/13),z=2^(1/13)*%e^(-(2*%i*%pi)/13), z=2^(1/13),z=0] (%i3) multiplicities; (%o3) [1,1,1,1,1,1,1,1,1,1,1,1,1,1] (%i4) z2:z1^14-z1; (%o4) (z^14-z)^14-z^14+z (%i5) solve(z2=z); (%o5) [z=2^(1/13)*%e^((2*%i*%pi)/13),z=2^(1/13)*%e^((4*%i*%pi)/13), z=2^(1/13)*%e^((6*%i*%pi)/13),z=2^(1/13)*%e^((8*%i*%pi)/13), z=2^(1/13)*%e^((10*%i*%pi)/13),z=2^(1/13)*%e^((12*%i*%pi)/13), z=2^(1/13)*%e^(-(12*%i*%pi)/13),z=2^(1/13)*%e^(-(10*%i*%pi)/13), z=2^(1/13)*%e^(-(8*%i*%pi)/13),z=2^(1/13)*%e^(-(6*%i*%pi)/13), z=2^(1/13)*%e^(-(4*%i*%pi)/13),z=2^(1/13)*%e^(-(2*%i*%pi)/13), z=2^(1/13),z=0,0=z^78-7*z^65+21*z^52-35*z^39+35*z^26-21*z^13+7, 0=z^78-5*z^65+11*z^52-13*z^39+9*z^26-3*z^13+1] (%i6) multiplicities; (%o6) [1,1,1,1,1,1,1,1,1,1,1,1,1,27,1,1] z^15 +m*z dynamic plane z^15-z thumbnail|right|Julia set for f(z)= z^15-z thumbnail|right|Critical points and orbits for f(z)= z^15-z How to compute iterations : /* Maxima CAS session */ (%i1) z:x+y*%i; (%o1) %i*y+x (%i2) z1:z^15-z; (%o2) (%i*y+x)^15-%i*y-x (%i3) realpart(z1); (%o3) -15*x*y^14+455*x^3*y^12-3003*x^5*y^10+6435*x^7*y^8-5005*x^9*y^6+1365*x^11*y^4-105*x^13*y^2+x^15-x (%i4) imagpart(z1); (%o4) -y^15+105*x^2*y^13-1365*x^4*y^11+5005*x^6*y^9-6435*x^8*y^7+3003*x^10*y^5-455*x^12*y^3+15*x^14*y-y Critical points : (%i1) m:-1; f:z^15+ m*z; d:diff(f,z,1); s:solve(d=0,z)$ s:map(rhs,s)$ s:map(rectform,s)$ s:map('float,s); multiplicities; (%o1) -1 (%o2) z^15-z (%o3) 15*z^14-1 (%o7) [0.35757475986465*%i+0.74251163973317, 0.64432745317147*%i+0.51383399763062, 0.80346319222004*%i+0.1833852305369, 0.80346319222004*%i-0.1833852305369, 0.64432745317147*%i-0.51383399763062, 0.35757475986465*%i-0.74251163973317, -0.8241257452789, -0.35757475986465*%i-0.74251163973317, -0.64432745317147*%i-0.51383399763062, -0.80346319222004*%i-0.1833852305369, 0.1833852305369-0.80346319222004*%i, 0.51383399763062-0.64432745317147*%i, 0.74251163973317-0.35757475986465*%i, 0.8241257452789] (%o8) [1,1,1,1,1,1,1,1,1,1,1,1,1,1] It means that here are 14 critical points and 14 critical orbits. Fixed points : kill(all); remvalue(all); /*------------- functions definitions ---------*/ /* function */ f(z):=z^15 -z; /* find fixed points returns a list */ GiveFixedPoints():= block ( [s], s:solve(f(z)=z), /* remove "z=" from list s */ s:map('rhs,s), s:map('rectform,s), s:map('float,s), return(s) )$ compile(all); ff:GiveFixedPoints(); multiplicities; length(s); for i:1 thru length(ff) step 1 do (z:ff[i], disp("z= ",z, " abs(d(z))= ",abs(15*z^14-1))); Result is : (%i12) ff:GiveFixedPoints() (%o12) [0.45590621928146*%i+0.94669901916834,0.82151462051137*%i+0.65513604843564,1.024411975933374*%i+0.23381534859391, 1.024411975933374*%i-0.23381534859391,0.82151462051137*%i-0.65513604843564,0.45590621928146*%i-0.94669901916834,-1.050756638653219,- 0.45590621928146*%i-0.94669901916834,-0.82151462051137*%i-0.65513604843564,-1.024411975933374*%i-0.23381534859391,0.23381534859391- 1.024411975933374*%i,0.65513604843564-0.82151462051137*%i,0.94669901916834-0.45590621928146*%i,1.050756638653219,0.0] (%i13) multiplicities (%o13) [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1] (%i14) length(s) (%o14) 14 (%i15) for i thru length(ff) do (z:ff[i],disp("z= ",z," abs(d(z))= ",abs(15*z^14-1))) z= 0.45590621928146*%i+0.94669901916834 ; abs(d(z))= 28.99999999999996 z= 0.82151462051137*%i+0.65513604843564 ; abs(d(z))= 28.99999999999998 z= 1.024411975933374*%i+0.23381534859391 ; abs(d(z))= 28.99999999999999 z= 1.024411975933374*%i-0.23381534859391 ; abs(d(z))= 28.99999999999997 z= 0.82151462051137*%i-0.65513604843564 ; abs(d(z))= 29.00000000000001 z= 0.45590621928146*%i-0.94669901916834 ; abs(d(z))= 28.99999999999995 z= -1.050756638653219 ; abs(d(z))= 29.00000000000003 z= -0.45590621928146*%i-0.94669901916834; abs(d(z))= 28.99999999999995 z= -0.82151462051137*%i-0.65513604843564; abs(d(z))= 29.00000000000001 z= -1.024411975933374*%i-0.23381534859391 ; abs(d(z))= 28.99999999999997 z= 0.23381534859391-1.024411975933374*%i abs(d(z))= 28.99999999999999 z= 0.65513604843564-0.82151462051137*%i ; abs(d(z))= 28.99999999999998 z= 0.94669901916834-0.45590621928146*%i ; abs(d(z))= 28.99999999999996 z= 1.050756638653219 ; abs(d(z))= 29.00000000000003 z= 0.0 ; abs(d(z))= 1.0 So only z=0 is parabolic fixed points, the rest of them are repelling Other multibrotmultibrot by Claude z^2 +cz^5z^2 +cz^5 z^5+m*z^4+z dynamica plane m = 08+0.8i thumb|right|Critical orbits f(z) =z^5 +(0.8+0.8)*z^4 + z Descriptionmath.stackexchange question: whats-the-difference-between-a-total-basin-of-attraction-and-an-immediate-basin map : coefficients in ascending order ( from ao to an): 0,1,0,0, 0.8+0.8i, 1 ListOfCriticalPoints [(- 0.7558074500261052 %i) - 0.7558074500261052, 0.2793534499540583 %i - 0.5310341598343944, 0.2793534499540583 - 0.5310341598343943 %i, 0.3674881599064412 %i + 0.3674881599064413] fixed points : [(- 0.8 %i) - 0.8, 0.0] with stability : [0.6384000000000008, 1.0] m=0.8+0.4i Inside dynamics under generalized power setting The orbit dynamics of the set can become more complex when the power is something other than 2.0. The iterated function can become multivalued and the structure of the set is then affected by the 'arbitrary' choice of which value is chosen. See also :commons:Category:Complex polynomial maps Derivative of iterated fuction Mark McClure : visualization of polynomial_julia_sets, coefficients a0, a1, a2, a3, .... for f(z) = an*z^n+ ... + a2*z^2 + a1*z + a0, example : 1.42+0.37i,-1.5,0,1 Exploring the 4-Dimensional Mandelbrot Set Elliptic curve julia sets by Clifford A. Reiter FF: Julia sets: True shape and escape time References