function [H, HI, D1, D2, D4, e_l, e_r, M4, d2_l, d2_r, d3_l, d3_r, d1_l, d1_r] = d4_variable_6_3(m,h)
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    %%% 6:te ordn. SBP Finita differens         %%%
    %%% operatorer med diagonal norm            %%%
    %%% Extension to variable koeff             %%%
    %%%                                         %%%
    %%% H           (Normen)                    %%%
    %%% D1=H^(-1)Q  (approx f?rsta derivatan)   %%%
    %%% D2          (approx andra derivatan)    %%%
    %%% D2=HI*(R+C*D*S                          %%%
    %%%                                         %%%
    %%% R=-D1'*H*C*D1-RR                        %%%
    %%%                                         %%%
    %%% RR ?r dissipation)                      %%%
    %%% Dissipationen uppbyggd av D4:           %%%
    %%% DI=D4*B*H*D4                            %%%
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

    % H?r med 7 RP ist?llet f?r 8 f?r D4 operatorn, dock samma randderivator
    % Denna ?r noggrannare, och har 2a ordningens randdslutning och b?r ge 6te
    % ordningens konvergens. Hade 2 fria parametrar att optimera



    H=diag(ones(m,1),0);
    H(1:7,1:7)=[
        0.414837907e9/0.1191965760e10 0 0 0 0 0 0;
        0 0.475278367e9/0.397321920e9 0 0 0 0 0;
        0 0 0.13872751e8/0.12416310e8 0 0 0 0;
        0 0 0 0.346739027e9/0.595982880e9 0 0 0;
        0 0 0 0 0.560227469e9/0.397321920e9 0 0;
        0 0 0 0 0 0.322971631e9/0.397321920e9 0;
        0 0 0 0 0 0 0.616122491e9/0.595982880e9;
    ];

    H(m-6:m,m-6:m) = fliplr(flipud(H(1:7,1:7)));


    e_1=zeros(m,1);
    e_1(1)=1;
    e_m=zeros(m,1);
    e_m(m)=1;

    S_U=[-0.137e3/0.60e2 5 -5 0.10e2/0.3e1 -0.5e1/0.4e1 0.1e1/0.5e1;]/h;
    S_1=zeros(1,m);
    S_1(1:6)=S_U;
    S_m=zeros(1,m);
    S_m(m-5:m)=fliplr(-S_U);


    S2_U = [0.15e2/0.4e1 -0.77e2/0.6e1 0.107e3/0.6e1 -13 0.61e2/0.12e2 -0.5e1/0.6e1;]/h^2;
    S2_1 = zeros(1,m);
    S2_1(1:6) = S2_U;
    S2_m = zeros(1,m);
    S2_m(m-5:m) = fliplr(S2_U);


    S3_U = [-0.17e2/0.4e1 0.71e2/0.4e1 -0.59e2/0.2e1 0.49e2/0.2e1 -0.41e2/0.4e1 0.7e1/0.4e1;]/h^3;
    S3_1 = zeros(1,m);
    S3_1(1:6) = S3_U;
    S3_m = zeros(1,m);
    S3_m(m-5:m) = fliplr(-S3_U);

    %DS=zeros(m,m);
    %DS(1,1:5)=-[-25/12, 4, -3, 4/3, -1/4];
    %DS(m,m-4:m)=fliplr(-[-25/12, 4, -3, 4/3, -1/4]);
    %DS=diag(c)*DS/h;


    H=h*H;
    HI=inv(H);


    % Fourth derivative, 1th order accurate at first 8 boundary points (still
    % yield 5th order convergence if stable: for example u_tt=-u_xxxx

    m4 = 7/240;
    m3 = -2/5;
    m2 = 169/60;
    m1 = -122/15;
    m0 = 91/8;

    M4 = m4*(diag(ones(m-4,1),4)+diag(ones(m-4,1),-4))+m3*(diag(ones(m-3,1),3)+diag(ones(m-3,1),-3))+m2*(diag(ones(m-2,1),2)+diag(ones(m-2,1),-2))+m1*(diag(ones(m-1,1),1)+diag(ones(m-1,1),-1))+m0*diag(ones(m,1),0);

    %M4=(-1/6*(diag(ones(m-3,1),3)+diag(ones(m-3,1),-3) ) + 2*(diag(ones(m-2,1),2)+diag(ones(m-2,1),-2)) -13/2*(diag(ones(m-1,1),1)+diag(ones(m-1,1),-1)) + 28/3*diag(ones(m,1),0));

    M4_U = [
        0.1399708478939e13/0.263487168000e12 -0.13482796013041e14/0.834376032000e12 0.344344095859e12/0.17565811200e11 -0.3166261424681e13/0.250312809600e12 0.1508605165681e13/0.333750412800e12 -0.486270829441e12/0.834376032000e12 -0.221976356359e12/0.5006256192000e13;
        -0.13482796013041e14/0.834376032000e12 0.7260475818391e13/0.139062672000e12 -0.27224036353e11/0.406022400e9 0.1847477458951e13/0.41718801600e11 -0.848984558161e12/0.55625068800e11 0.247494925991e12/0.139062672000e12 0.165585445559e12/0.834376032000e12;
        0.344344095859e12/0.17565811200e11 -0.27224036353e11/0.406022400e9 0.2044938640393e13/0.22250027520e11 -0.1071086785417e13/0.16687520640e11 0.502199537033e12/0.22250027520e11 -0.143589154441e12/0.55625068800e11 -0.88181965559e11/0.333750412800e12;
        -0.3166261424681e13/0.250312809600e12 0.1847477458951e13/0.41718801600e11 -0.1071086785417e13/0.16687520640e11 0.628860435593e12/0.12515640480e11 -0.73736245829e11/0.3337504128e10 0.195760572271e12/0.41718801600e11 -0.81156046361e11/0.250312809600e12;
        0.1508605165681e13/0.333750412800e12 -0.848984558161e12/0.55625068800e11 0.502199537033e12/0.22250027520e11 -0.73736245829e11/0.3337504128e10 0.76725285869e11/0.4450005504e10 -0.3912429433e10/0.406022400e9 0.53227370659e11/0.17565811200e11;
        -0.486270829441e12/0.834376032000e12 0.247494925991e12/0.139062672000e12 -0.143589154441e12/0.55625068800e11 0.195760572271e12/0.41718801600e11 -0.3912429433e10/0.406022400e9 0.1699707221791e13/0.139062672000e12 -0.6959018412841e13/0.834376032000e12;
        -0.221976356359e12/0.5006256192000e13 0.165585445559e12/0.834376032000e12 -0.88181965559e11/0.333750412800e12 -0.81156046361e11/0.250312809600e12 0.53227370659e11/0.17565811200e11 -0.6959018412841e13/0.834376032000e12 0.3012195053939e13/0.263487168000e12;
    ];

    M4(1:7,1:7) = M4_U;

    M4(m-6:m,m-6:m) = flipud( fliplr( M4_U ) );
    M4 = M4/h^3;

    D4 = HI*(M4-e_1*S3_1+e_m*S3_m  + S_1'*S2_1-S_m'*S2_m);
end