sqp method - slow

Hi,

I run the sqp, active-set and interior-point methods on the same set of data and their options are the same. I am confused why it takes 20 minutes for sqp to find the solution while for the other two methods it takes less than 5 seconds.

My goal function is:

Err = sum((F_observed - F_raw) .^2);

Here is the 'iter' output:

                                                            Norm of First-order
   Iter F-count            f(x) Feasibility  Steplength        step  optimality
      0      16    2.027691e+05   0.000e+00                           2.611e+06
      1      55    2.015284e+05   0.000e+00   2.737e-04   8.211e-03   7.520e+06
      2      88    1.915127e+05   0.000e+00   2.326e-03   6.598e-02   2.434e+06
      3     122    1.892994e+05   0.000e+00   1.628e-03   4.342e-02   1.879e+06
      4     160    1.876097e+05   0.000e+00   3.910e-04   1.067e-02   4.279e+06
      5     189    1.785387e+05   0.000e+00   9.689e-03   2.448e-01   5.118e+06
      6     218    1.612519e+05   0.000e+00   9.689e-03   2.418e-01   3.995e+06
      7     244    1.543721e+05   0.000e+00   2.825e-02   6.225e-01   3.297e+06
      8     271    1.370354e+05   0.000e+00   1.977e-02   4.613e-01   3.018e+06
      9     295    1.306121e+05   1.000e+00   5.765e-02   1.114e+00   4.053e+06
     10     324    1.243357e+05   1.000e+00   9.689e-03   2.222e-01   2.623e+06
     11     348    1.210242e+05   1.000e+00   5.765e-02   1.063e+00   4.090e+06
     12     372    1.083141e+05   0.000e+00   5.765e-02   1.042e+00   2.634e+06
     13     393    1.041635e+05   0.000e+00   1.681e-01   1.947e+00   3.196e+06
     14     417    8.626112e+04   1.000e+00   5.765e-02   8.158e-01   1.418e+06
     15     438    8.510665e+04   0.000e+00   1.681e-01   1.190e+00   2.006e+06
     16     455    7.392513e+04   0.000e+00   7.000e-01   3.826e+00   1.067e+06
     17     471    6.790133e+04   0.000e+00   1.000e+00   1.544e+00   3.717e+05
     18     496    6.783769e+04   0.000e+00   4.035e-02   8.082e-01   3.947e+05
     19     512    6.718949e+04   0.000e+00   1.000e+00   8.408e-01   7.938e+04
     20     528    6.704678e+04   0.000e+00   1.000e+00   1.287e-01   3.632e+04
     21     544    6.701860e+04   0.000e+00   1.000e+00   2.931e-02   2.684e+04
     22     560    6.700750e+04   0.000e+00   1.000e+00   5.487e-02   2.600e+04
     23     576    6.700702e+04   0.000e+00   1.000e+00   2.736e-03   1.835e+03
     24     592    6.700700e+04   0.000e+00   1.000e+00   1.416e-03   1.034e+03
     25     608    6.700700e+04   0.000e+00   1.000e+00   2.349e-04   3.498e+01
     26     624    6.700700e+04   0.000e+00   1.000e+00   2.100e-05   6.189e+00
     27     640    6.700700e+04   0.000e+00   1.000e+00   6.972e-06   1.251e+00
     28     641    6.700700e+04   0.000e+00   7.000e-01   8.553e-07   4.189e-01

Optimization completed: The relative first-order optimality measure, 6.074215e-07,
is less than options.TolFun = 1.000000e-06, and the relative maximum constraint
violation, 0.000000e+00, is less than options.TolCon = 1.000000e-06.

Optimization Metric                                         Options
relative first-order optimality =   6.07e-07       TolFun =   1e-06 (selected)
relative max(constraint violation) =   0.00e+00    TolCon =   1e-06 (selected)