Open rdemaria opened 5 years ago
@rdemaria Are you sure this worked in the past? I couldn't find a version it works for and it doesn't look like it was ever considered in the current implementation?
The last I found is /afs/cern.ch/user/m/mad/madx/releases/old/V3.02/madx
which works with this example:
TITLE,'Transferline matching with MAD-X';
option,-info,-warn;
call file="fv9.opt";
call file="fv9.seq";
option,info,warn;
kqf = 0.0087;
kqd = -0.0091;
Beam, particle = proton, sequence=fivecell, energy = 450.0,
NPART=1.05E11, sige= 4.5e-4 ;
use,period=fivecell;
savebeta,label=abc,place=#s;
select,flag=twiss,range=#s/#e,column=name,s,betx,bety;
twiss, sequence=fivecell;
match,sequence=fivecell,beta0=abc,x=0,px=0,y=0,py=0,t=0,pt=0;
constraint,sequence=fivecell,range=#e,betx=180.0,bety=30.0;
vary,name=kqf,step=1.0e-6;
vary,name=kqd,step=1.0e-6;
jacobian,calls=10,tolerance=1.e-15,strategy=2;
!lmdif,calls=500,tolerance=1.e-15;
endmatch;
JACOBIAN, STRATEGY=2
print the response matrix (jacobian) of matching problem and write a report based on the SVD. This works with USE_MACRO and it used to work in normal matching few years ago, but now it crashes madx with a segfault.