Dear Axel,
attached to this post is a plot of the present cluster expansion. Obviously the ground states appear to be partially wrong as some of them lie on top of each other. Can you help me resolve this problem? The ground state search has been restricted with the crange.in to the concentration range 0.5<c<1.
Thanks. Siau
Is this for a binary alloy or more than 2 components?
If the latter, this explains the problem: use mmapsrep instead of mapsrep.
Of course, with higher dimensions, viewing capability are more limited!
Ah, yes. This has more than 2 components.
However when I use mmapsrep it terminates with
splot ‘./gs_lines.out’ u (0.5*(2*+)/(++)):(sqrt(3)/2*()/(++$))
8) t "known gs" w lp pt 4, './fit.out' u (0.5*(2*+)/(++)):(sqrt(3)/2*()/(++))8) t "known str" w p pt 1, './newgs.out' u (0.5*(2*+)/(++)):(sqrt(3)/2*()/(++))7) t "predicted gs" w p pt 3, './predstr.out' u (0.5*(2*+)/(++)):(sqrt(3)/2*()/(++)):($7) t "predicted" w p pt 0
^
"mapsrep.gnu", line 12: Column number expected
Error running gnuplot or …
You do not have gnuplot installed.
So it has to be an error, since I have gnuplot 4.6 installed. Do I perhaps need a newer version?
OK, so I installed the newest version of gnuplot. Unfortunately the problem remains.
Do I assume correctly that the mmapsrep script was written so that the resulting mapsrep.gnu has to be adapted by the user to his own needs?
Sorry, mmapsrep needs a few arguments:
mmapsrep -h gives you help:
Thanks. It still seems to show too many lines. It is not possible to make out the convex hull since all of the ground state points are interconnected when gs_lines.out is plotted.
How many components do you have? I’d need your gs* files to see what the problem is.
These are my ground states in gs.out:
0.000000 0.166667 0.083333 0.250000 0.500000 0.000000 0.339175 0.396854 -0.057679 7
0.000000 0.000000 0.250000 0.250000 0.500000 0.000000 0.000000 -0.006118 0.006118 0
0.250000 0.000000 -0.000000 0.250000 0.500000 0.000000 -0.000000 0.005372 -0.005372 8
0.041667 0.041667 0.166667 0.208333 0.500000 0.041667 -0.181237 -0.096728 -0.084509 199
0.000000 0.000000 0.250000 0.208333 0.500000 0.041667 -0.163492 -0.157201 -0.006291 70
0.000000 0.000000 0.250000 0.166667 0.500000 0.083333 -0.322402 -0.308284 -0.014117 106
0.000000 0.041667 0.208333 0.208333 0.500000 0.041667 -0.173244 -0.142559 -0.030685 97
0.041667 0.083333 0.125000 0.125000 0.500000 0.125000 -0.449967 -0.422979 -0.026988 162
0.000000 0.041667 0.208333 0.166667 0.500000 0.083333 -0.331070 -0.293642 -0.037428 136
0.041667 0.041667 0.166667 0.166667 0.500000 0.083333 -0.339775 -0.247811 -0.091964 224
0.000000 0.041667 0.208333 0.125000 0.500000 0.125000 -0.464444 -0.455185 -0.009259 154
0.000000 0.000000 0.250000 0.125000 0.500000 0.125000 -0.473134 -0.459368 -0.013766 148
0.041667 0.041667 0.166667 0.125000 0.500000 0.125000 -0.468386 -0.412317 -0.056068 155
0.000000 0.250000 0.000000 0.000000 0.500000 0.250000 0.000000 -0.016653 0.016653 2
0.000000 0.083333 0.166667 0.000000 0.500000 0.250000 -0.322735 -0.294107 -0.028628 50
0.083333 0.000000 0.166667 0.000000 0.500000 0.250000 -0.404800 -0.313630 -0.091169 49
0.125000 0.000000 0.125000 0.125000 0.500000 0.125000 -0.330067 -0.349993 0.019925 158
0.041667 0.000000 0.208333 0.125000 0.500000 0.125000 -0.446069 -0.416500 -0.029570 149
0.000000 0.000000 0.250000 0.000000 0.500000 0.250000 -0.203597 -0.432834 0.229237 48
I have also attached gs_lines.out.
You have 5 independent compositions! How is the code supposed to plot that on a 2d screen?!
Sorry I have no solution to this.
Are c1~c3 the three dimensions in the fig? What if I only wanted to have two variables and leave the third axis for energy?
It will generate a 2d point from the three composition (yes I know, one is redundant). The 3rd axis will be for energy.
BTW, the command is mmapsrep (2 m’s - I will correct the help file).