[MPB-discuss] Fw: Calculating h-fields from the eigenvectors (check)

[r.maspero]

?Hi,


I am running a check in MPB to make sure that I understand what the 
eigenvectors, I have output, are. I figured the easiest check was to compare 
the h-fields generated by MPB to the h-fields that I recreated using the 
eigenvectors that I output from the same simulation.

I have attached the matlab code (quick2.m) that I wrote to calculate the fields 
using the eigenvectors where "eigenvecs.dat" is generated by using "h5totxt" on 
the eigenvector output. I have also attached the .ctl file (firstbs.ctl)

However, the fields I generate are completely different to the fields output by 
MPB. Is there something I am misunderstanding in what the eigenvectors are, how 
to make the G vectors or is there a trick you do that I have missed?

Thank you

Ross

firstbs.ctl
Description: firstbs.ctl

clear all;
load('eigenvecs.dat');

a = [1 0; 0 1];
b = [2*pi 0; 0 2*pi];

for i=1:size(eigenvecs,1)/2
    temp(i,:) = complex(eigenvecs(2*i-1,:), eigenvecs(2*i,:));
    EV(ceil(i/2),abs(mod(i,2)-2),:) = temp(i,:);
end

res=floor(sqrt(size(EV,1))); count = 0;
for i=1:res
    for j=1:res
        count = count + 1;
        r(count,1) = i/res;
        r(count,2) = j/res;
    end
end

for i=1:size(r,1)
    G(i,:) = b(1,:)*r(i,1) + b(2,:)*r(i,2);
end


for i=1:size(r,1)
    E(i) = 0;
    for j=1:size(G, 1)
        E(i) = E(i) + EV(j,1,2)*exp(1i*dot(G(j,:), r(i,:)));
    end
end

count = 1;
for i=1:res
    for j=1:res
        results(i,j) = E(count);
        count = count + 1;
    end
end


figure;
surf(real(results), 'EdgeColor','None');
view(2);
        
    
_______________________________________________
mpb-discuss mailing list
mpb-discuss@ab-initio.mit.edu
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/mpb-discuss