about the optimization

[cgwang]

Dear all,
I used IOPT = 7 (fire) to optimize the initial and final states of NEB. But after more than 300 steps, they didn't converge. The EDIFFG=-0.02, and I used RPBE+D. At the end of calculation, the stand out file is this:
RMM: 3 -0.497035870373E+03 0.29104E-10 0.00000E+00 5 0.678E-04 0.409E-09
RMM: 4 -0.497035870373E+03 0.43656E-09 0.00000E+00 5 0.678E-04
150 F= -.53507652E+03 E0= -.53508496E+03 d E =0.640284E-09
quench: g(F)= 0.000E+00 g(S)= 0.000E+00 dE (1.order)= 0.000E+00
bond charge predicted
N E dE d eps ncg rms rms(c)
RMM: 1 -0.497035870373E+03 -0.32014E-09 0.00000E+00 5 0.678E-04 0.627E-09
RMM: 2 -0.497035870373E+03 0.90222E-09 0.00000E+00 5 0.678E-04 0.616E-09
RMM: 3 -0.497035870372E+03 0.26193E-09 0.00000E+00 5 0.678E-04 0.573E-09
RMM: 4 -0.497035870372E+03 0.40745E-09 0.00000E+00 5 0.678E-04

The dE is very small. But it still didn't converge. I'm confused by this. Is there something wrong with my compiling of VASP? Or it dues to somthing else? Looking forward for your reply!

Best regards

[cgwang]

I have got what leads to the results. The tag 'IOPT = 7' in the INCAR makes no sense to the calculation. The tag 'IOPT' didn't work in VASP. It must be some wrong with the compiling of VASP. Do you have any ideas on how to optimizie the compiling? Thank you!

[graeme]

The IOPT tags are available if you have linked in our vtstcode into vasp:
http://theory.cm.utexas.edu/vasp/downloads/

You can grep for "VTST" in the OUTCAR to see if they were included in your build.

[graeme]

This issue should be fixed now.